This PDF file contains the front matter associated with SPIE Proceedings Volume 12703 including the Title Page, Copyright information, Table of Contents, and Conference Committee Page.

The ramp taxiway is an important area for aircraft to enter and exit the aircraft space. This paper firstly introduces different numbers of ramp taxiway operation modes, then summarizes their advantages and disadvantages and conducts a comparative analysis. Taking the design of the new terminal area of Nanning Airport as an example, Simmod software is applied to simulate and evaluate three different design and planning schemes. Therefore, the three-lane unidirectional operation scheme is chosen as the final setting layout scheme.

Compared with the traditional separated source-channel coding model, the performance of the joint source-channel coding will be better in the actual communication system. In addition, the typical decoding algorithms all belong to the algorithms of separated source-channel model. Therefore, the joint coding and decoding algorithm is particularly important and it will be recognized as of great significance in future communication. In this paper, in order to meet the future communication requirements, a scheme of polar code in joint coding is proposed. At the transmitter of the system, the source polarization is constructed by the Bhattacharyya parameter construction method and encoded by non- systematic polar code. For channel polarization, polarization weight construction method is used, which was independent of signal to noise ratio. A joint Belief Propagation (BP) decoder with a simple structure is adopted in the receiver, which includes a channel decoder and a source decoder. The joint decoder performs external iteration only through a few iterations, thus reducing the complexity and improving the efficiency of the communication system. Simulation results show that compared with the channel BP decoder, the performance gain of J-BP decoder is about 1.8dB.

China is one of the world's largest producers of lithium batteries, and with the widespread use of batteries, the transportation of lithium battery cargoes has attracted great attention. The UN38.3 standard, as a safety test basis for the transportation of lithium batteries widely used worldwide, aims to clarify the requirements for the transportation of lithium battery cargo and to ensure its safe transportation by air and land cargo. This paper will introduce the importance of lithium battery testing and various testing items, and discuss the practicalities of using the UN38.3 standard as a basis for analysing the practice of testing in the manufacturing industry and discussing how to use automated and intelligent technology to carry out testing and testing processes for reference in the manufacturing industry.

The new power system has a new operation mode of "the source follows the load, the load follows the network, and the source loads interact". With the continuous excavation of flexible regulation resources, new third-party entities such as virtual power plants or load integrators need to consider implementing differentiated network security protection requirements to ensure their fair access to business. At the same time, with the clarity of the regulation level of the new power system and the continuous participation of various entities in the construction, the company's network security protection boundary will no longer be "horizontal and vertical", but will become more uneven, with more types of interfaces and interfaces. It is necessary to refine the boundary protection measures according to different interaction needs. The new regulatory subject on the demand side of the new power system involves the privacy information of a large number of users. Data flow and sharing, cross access and collaborative analysis among the subjects have security problems such as unauthorized access and privacy leakage. This paper focuses on the scenario based intelligent desensitization and dynamic access control technology for power distribution and distributed new energy data resources, and combs various factors involved in the data access process, On the basis of obtaining the data access authority, the dynamic correlation analysis of the three elements of user role and access authority, sensitive data level, power allocation and distributed new energy business scenario is integrated, and finally the appropriate desensitization method is selected from the desensitization algorithm library, and finally the appropriate desensitization method is selected from the desensitization algorithm library to ensure the security protection capability of the new distribution network under the flexible, open and interactive energy scenario, Effectively respond to new major network security risks.

The traceability and analysis of power distribution business data depend on the accurate description of the correlation network of power distribution data resources. In response to the above requirements, this paper first builds a multi-source heterogeneous power distribution and distributed new energy data flow topology network. Aiming at the complex relationship between power distribution and data resources and the flow characteristics of data, topology network construction technology is used to build bus type, star type, ring type, tree type, mesh and other topology networks according to the data association characteristics to achieve accurate description of the data association network. Then, based on the data association relationship network of power distribution and distributed energy, the data traceability must be calculated, and the traceability analysis can only be carried out if there is connectivity between the data. In response to the above requirements, the research on connectivity analysis algorithm based on power distribution and distributed new energy data network was carried out. By solving the connection path between the data nodes, the connectivity analysis requirements of the data nodes in the directed connected network and the directed connected network can be met. The data traceability of power distribution and distributed new energy is to trace the source of data in all business links, which plays an important role in the safe production, problem tracing, root cause analysis and other businesses of power distribution. In response to the above requirements, carry out research on tracing technology of allocated electricity and distributed new energy

The application of electric distribution in various fields of society is very common and indispensable. It has become one of the most important indicators to measure a country's comprehensive national strength. The level of electric service has a very important impact on the production and quality of life of the whole society. With the continuous expansion of the scope of electric index analysis, the traditional electric index analysis tools can no longer meet the needs of massive electric big data analysis, and the rise of big data technology provides an effective solution. This paper first introduces the definition, measurement indicators and common theoretical analysis methods of electric distribution data analysis. Based on the introduction of electric data analysis technology, it analyzes the development trend of electric quality analysis in the big data environment. Then, according to the characteristics of electric data of distribution network, an electric quality big data storage scheme based on Mongo DB is proposed, a power big data processing flow is designed, and a electric big data computing framework based on apache spark is built. On the basis of completing the construction of electric big data of distribution network, this paper expounds the overall design idea, system framework design and system function module design of electric data analysis system and gives the design scheme of electric quality analysis system based on B/S four tier architecture.

Today's society is in the era of rapid development of science and technology. With the continuous updating and iteration of artificial intelligence science, visual art has ushered in a new chapter of digitalization and intelligence with the help of powerful computer and network technology. Modern painting is a kind of painting form that gradually matures in the period of extreme development of industrial society. The anti-aesthetic trend of thought embodied in modern painting is a recognized judgment of modern painting. Therefore, this paper will focus on the possibility of diversified expression of visual arts with the help of artificial intelligence related technologies. The art talent education system, newspaper and magazine communication system, art association system, and art criticism system possessed by modern Chinese art all constitute an important part of the art system. The guidance and restriction of the art system enable Chinese art to transcend the imagination and monologue of the individual soul in literati paintings, move towards the value orientation of life and socialization, and form a new realistic tradition of realism, life and popularity. This paper selects a technical branch of the field of artificial intelligence - the field of deep learning. Combining with the manifestation of visual art, it explores the characteristics of intelligent visual art in the context of deep learning, finds its advantages that can be combined with artistic expression, and creates a new form of visual art creation expression based on deep learning.

As an important guide for grid development and a key link of grid construction, distribution network planning should not only adapt to the trend of intelligence and refinement in grid development and construction, but also meet the needs of users for reliable power consumption. The quality of electric distribution planning scheme exerts significant influence on the capacity of electric supply, operation level, economic measures of electric grid. Depending on the background that power supply enterprises vigorously promote the regional grid division of electric distribution at this stage, the regional grid division of electric distribution is comprehensively evaluated by establishing a scientific and effective indicator set of electric consumption management, which has important practical significance in analyzing the associated defects in the operation process of electric grid and determining the effectiveness of electric grid planning. In order to comprehensively and effectively evaluate the remote electric grid indicator set. In this article, considering the five important influencing factors of urban electric grid namely, capacity of electric supply, structure of electric grid, quality of electric supply, standard of device and investment benefit, and establishes a scientific and perfect indicator label orient to the objects and index levels involved in the evaluation and the rationality of electric consumption management. The indicator set of electric consumption management of distribution grid planning with clear hierarchy is proposed, and the definition and calculation method of the indicator are proposed.

The original data center management operation and maintenance is usually assisted by the data center environmental monitoring system. All along, the environmental monitoring system only collects and displays the running data of the testing equipment and does not participate in the equipment configuration and automatic control. Artificial Intelligence (AI), as a tool to explore and find the best management and operation and maintenance strategy, will make the intricate linkage and coordination of mechanical and electrical equipment in data centers easier to realize. This paper reveals the inevitability of the birth of AI data center and proposes a large data center security transmission model based on AI under the background of digital transformation, which reduces the resource consumption of protection routing and improves the survivability of data center network. The simulation results show that the number and link length of protection routes established by this algorithm are less. On the basis of ensuring that the data center is not affected by the failed nodes and the data is transmitted normally and effectively, the resource consumption of protection routes is reduced, and the survivability of the data center network is improved.

The power system applies high precision satellite navigation signal as a unified time reference to realize the power network scheduling control. Chinese electric power system mainly adopts the space-based time-timing scheme based on Bei Dou navigation Satellite (BDS) and assisted by Global Position System (GPS). However, due to the semi-open character of civil GNSS applications, the risk of interference to the time synchronization of the power system is caused by various kinds of radio waves, and it leads to synchronization network disorder, and affects the stableness and reliability of the operating electric power grid. This paper analyzes the principle of satellite timing synchronization in power system and the characteristic of interference signals, puts forward a targeted design scheme of satellite time synchronization monitoring system, and analyzes the key technologies, and to provide technical reference and improvement ideas for the safety protection of power time synchronization system.

The polarity effect caused by the asymmetry and non-uniformity of electric field is one of the important basic characteristics of gas discharge, and the previous knowledge of polarity effect is mainly the differences of discharge initial voltage and electrical pulse statistical characteristics, while the lack of knowledge of the light emission characteristics and the “polarity effect” of photoelectric pulse accompanying the discharge. To better apply the optical detection and arc light monitoring in the SF₆ gas insulated system and further improve the understanding of the photoelectric pulse polarity effect, this paper builds a set of SF₆ partial discharge photoelectric simulation experimental system, and analyzes the multi-spectral photoelectric polarity effect by collecting the multi-spectral signals of insulator creepage and corona discharge with multi-spectral sensor. The research shows that there are significant differences of the amplitude and proportion of two type of discharges, among which the multispectral polarity effect of corona discharge is more obvious, with the rise of the applied voltage, the amplitude of multi-spectral signal along the surface discharge increases and the rate accelerates, while the rate of corona discharge slows down, and the proportion of UV band decreases for both discharge types, the proportion of visible band increases, and the infrared band remains low.

The economic profit of investment in energy storage systems are investigated with a regional-type grid as the research object. Firstly, the economic operation model of power supply and Energy Storage System (ESS) within the local grid is established, and the optimization model is solved by using hybrid particle swarm algorithm based on heuristic adjustment strategy. In order to reasonably assess the feasibility of the investment scheme, the equivalent net benefit method is designed to discount the economic benefits and investment costs of the ESS to a dispatch cycle based on the discount rate of life of the ESS and cost-benefit analysis. The algorithm gives the equilibrium return mean value point of ESS investment and performs a sensitivity analysis for the impact of investment unit price and price of electricity purchasing on the investment in ESS.

Filial piety is the natural expression of human beings' gratitude for their parents' fertility, upbringing and education. Confucian ethics is the source of Chinese traditional virtues and national spirit, and its core idea is "benevolence". Filial piety culture is the starting point and fundamental thought of "benevolence". It not only inherits the tradition of Chinese civilization, but also penetrates into the theoretical system of "benevolence", from the concept of natural consanguinity to the height of social morality and has formed a very far-reaching influence. With the development of social economy, consumers' personalized requirements for products are getting stronger and stronger, and their requirements for product performance and quality are getting higher and higher, which makes products more complicated. This paper analyzes the thinking form and creation characteristics of Confucian filial piety culture, finds out the semantic level and characteristics based on Confucian filial piety culture, and puts forward a product semantic level and characteristic analysis model based on deep learning. The results show that the overall performance of this model is better than that of the comparison model. Fusion of different levels of features can provide more context information for the final result, and the overall generalization performance of the model is good.

The electrolytic capacitor limits the service life of the LED driver and also renders the LED's long life worthless. The LED places a great demand on its associated driver's long life. This study suggests an electrolytic capacitor-free LED drive scheme to achieve high input power factor and flicker-free operation. Compared to existing electrolytic capacitor-less LED drivers, the proposed topology to eliminate the electrolytic capacitor and the circuit construction are simple. The operation principle, detailed design procedure of the main circuit, carried on the experimental verification last.

The automatic compliance review of substation design model greatly depends on the computer-understandable layout criteria. Unfortunately, most of these layout criteria are depicted in a series of design codes written in natural language. Meanwhile manual conversion from unstructured code text to layout rules is often costly and time consuming. In this regard, automatic extracting layout information can be developed utilizing Natural Language Processing (NLP) tools based on deep learning. The layout texts are collected and manually labeled to establish the experimental corpus for further study, and then the NLP-assisted information extraction method is developed mainly using the Bidirectional Long Short-Term Memory (BiLSTM) networks and Conditional Random Field (CRF). Finally, the information extraction model is trained on the labeled experimental corpus. The test results indicate that the developed model is feasible to extract semantic information of layout constraints residing in design code.

In this paper, a fault diagnosis scheme based on observer is proposed to solve the problem of servo motor fault in construction robot actuator system. Firstly, the mathematical model is established in abc coordinate system, and Park transformation is introduced to transform abc three-phase symmetry variable into dq coordinate DC through Park transformation, which greatly reduces the modeling complexity in normal state of motor, eliminate redundant motor models. Based on this, the state space equations are constructed, and an interval observer is designed. Compared with the traditional Luenberger interval observer, the interval observer proposed in this paper has better decoupling performance in solving the gain matrix, at the same time, it directly improves the robustness of the interval observer and the sensitivity to the fault signal under the electromagnetic interference. Finally, the method is verified on the fault simulation testbed, which proves the feasibility of the proposed method, and further improves the timeliness of fault detection of servo motor in construction robot actuator system.

Chopin founded a new school of piano playing, and enriched the piano playing skills with new artistic content and unprecedented performance. Because the data expression of music is a complex problem based on the multi-dimensional arrangement and combination of time series, based on the common characteristics of different styles of music, this paper compares the music score versions of Chopin etudes based on DL(Deep Learning), compares three performers and compares the sound versions, and finds that Chopin's melody is the most prominent in strength, with strong timbre. The bass melody of Boli version, Franç ois and Magalov is also outstanding, but the intensity is not so exaggerated. The results of this research are extended to the research and development of other composing system modules, which can effectively improve the efficiency and calculation results of machine DL. Obviously, there will be similarities and differences in the understanding and processing of the works, so it is very important to choose the version in piano teaching or performance.

Differential equation plays an important role in engineering technology, military, economy, medicine, biology, ecology and other fields. This paper uses differential equation to solve practical problems about carbon monoxide content. Through the analysis, modeling and solving of the case, not only the analytical solution of the equation is obtained, but also the numerical solution of the equation is given by using MATLAB program.

YOLOv3, as a multi-scale object detection algorithm, has a simple structure and can be detected quickly. However, during the training process, as the bottom convolutional layer contains more object detail, and the information in the underlying layer will be gradually faded out through continuous convolution. To address this, this paper adds a normalized attention mechanism behind the feature layers at different scales to further suppress insignificant features and enhance the discriminability and robustness of the features. Moreover, an improved pyramid structure is leveraged to fuse the detail of the shallow feature maps with the semantic feature information of the higher layers, aiming to improve the model performance. In particular, a YOLOv3 object detection algorithm named YOLO-MFE based on improved multi-scale feature extraction is proposed. Experimental results on the Pascal VOC dataset demonstrate that our algorithm outperforms the original YOLOv3 algorithm in terms of object detection accuracy, with a mAP on the Pascal VOC 2007 test set that is 1.66% higher.

In the LTE passive bistatic radar, the monitoring signal contains not only the main base station direct wave signal as well as multipath interference clutter, but also the direct wave signal and multipath interference signal from other co-channel base stations. When the detection environment is non-smooth and interference samples are not available, the conventional time domain filtering method cannot effectively suppress the clutter and co-channel interference. In order to solve this problem, this paper proposes a space-time cascaded spurious and co-channel interference suppression method. The method can not only effectively resist the effect of non-smooth clutter, but also suppress the co-channel interference suppression problem existing in LTE deployment stations. Theoretical analysis and simulation experiments verify the suppression performance of the proposed method for clutter interference.

Cyber Security plays a significant role in internet era nowadays, since information is exchanged all the time. During this period, the relationship between daily economic behavior and the network are gradually strengthened. Although network technology presents a trend of continuous development, there is still a long way to go before the technology becomes mature. In network applications, fraudulent websites commonly steal the private information of users. The implementation of data mining technologies to develop a deep learning model for identifying dangerous websites is a vital measure for network security.

A safe navigation route for USVs is designed to avoid obstacles when navigating in narrow waterway. A series of plane distribution points are selected in the navigation water area, and the adjacent distribution points are connected by a smooth curve, which is used to represent the obstacle avoidance path of USVs sailing in narrow waterway. The energy function of the length path and the energy of the collision penalty function for obstacles are defined as the total energy in paper for achieving the best avoidance path. With mathematical optimization methods, it is obtained that the total energy is extremely small so that each path point will move toward the direction of energy reduction, and that the optimal obstacle avoidance path is reached. The feasibility of the model is verified via computer simulation.

The drastic changes in the environment will cause a series of humanitarian disasters, lead to political violence, and make the already fragile country crumble. China's dependence on foreign trade remains high all the year round, and the development of international trade business is of great significance to China's economic development and the increase of economic aggregate.We should link the balance of international trade with the balance of international payments from the perspective of comparative advantage and economies of scale and rethink the dynamic balance of international trade. In this paper, the simulation research of international trade vector dynamic equilibrium model based on big data algorithm is carried out. We extend the estimation and inference method based on the dynamic equilibrium model of international trade vector to the uncertain region in the parameter space. Bayesian statistical inference method is used, and probability weights are constructed for deterministic and uncertain regions in parameter space based on observation data. The results show that the vector dynamic equilibrium model of international trade constructed in this paper can provide a reliable analytical framework and a powerful tool for policy makers to predict and judge the effect of monetary policy system changes.

Assessment of innovation ability is an important basis for higher vocational education and teaching decision-making, and effective assessment of innovation ability depends on comprehensive and reliable assessment data. Multi-assessment system of instructional level refers to a series of methods to assess teachers' teaching process and effect in unstructured teaching situation by using various effective technical means and assessment methods. With modern quality management theory, the contents and methods of multi-assessment are arranged and implemented by multi-assessment subjects. Based on the background of tertiary education and the comprehensive assessment of teachers' innovation ability in vocational schools, this article applies data mining technology to explore and discover the practical knowledge of teaching assessment in vocational schools. The results show that the consistency between the proposed algorithm and the manual processing method can reach more than 95%, that is, the accuracy of the algorithm is more than 95%, which shows that the assessment algorithm of teachers' innovation ability in this article has good performance. This algorithm reduces the influence of subjective factors in assessment, and makes the assessment results more objective and accurate, so it has strong application value.

SDG (Sustainable Development) is not only a concept, but also a development idea and a development model. The design orientation of China's foreign trade SDG evaluation index faces many choices. The SDG of foreign trade affects the SDG of economy and even the whole society. Realizing the SDG of foreign trade not only means promoting the implementation of the strategy of opening to the outside world, but also embodies the promotion and support for the strategy of SDG of national economy and society. Based on the analysis of big data, this paper constructs an evaluation index system for the SDG of foreign trade. The index setting should not only be guided by the theory and meet the inherent requirements of the theory, but also take objective data as the standard and take into account the data realizability. This paper makes a vertical and systematic analysis and comprehensive evaluation of trade development from five aspects: trade scale, trade benefit, economic benefit, ecological benefit and resource benefit. Reasonable and effective use of resources inside and outside the province, increase the total volume and scale of foreign trade in the province, optimize the foreign trade structure and industrial structure, and improve the efficiency of foreign trade, thus driving the sustained and healthy development of the province's economy.

At present, Internet technology is profoundly affecting various domestic industries, not only promoting industry innovation, but also changing various fields of economy and society. Educational services are no exception. They have also been influenced by Internet technology, and a new talent training model has been built on this basis. However, the current students' learning is still at a superficial stage, and their cognition is difficult to apply to practical problems, because a large amount of knowledge in the Internet environment is fragmented. Therefore, cultivating students' in-depth learning abilities, such as in-depth understanding of complex concepts, processing knowledge information, and constructing cognitive systems, is the real purpose of Internet + education. This paper will build a factor model that affects students' deep learning based on the collected information, and focus on solving the practical problems encountered by students in their learning as the main research direction, and explore the influencing factors of college students' deep learning in the Internet + environment.

The acquisition of load curves for typical days is a very important part in the study of load characteristics for power system, which is of great significance to the development of load dispatching plan or operation control. As the load characteristics are affected by various random factors, for instance, meteorology, human production and livelihood, etc., the selection of typical days, using traditional or single clustering algorithms, has some disadvantages to describe the overall load curve. To address those drawbacks, this paper proposes a new integrated clustering method by combining the improved Probabilistic Fuzzy C-Means (PFCM) algorithm with Fuzzy Linear Discriminant method (FLDA). Firstly, the original PFCM is improved to get the improved PFCM, and it is applied to the selection of optimal number of clusters. Then, the improved PFCM is combined with FLDA, and the integrated clustering algorithm is applied to the clustering of load curves. Ultimately, the annual load data of a power grid proves that the method has several certain advantages in the selection of typical days.

Multi-station integrated power grid system has a large number of cloud and edge data centers. It needs to solve the problem of collaborative use of cloud, edge and terminal resources, realize the rapid migration of computing tasks in the case of failure, and achieve the consistency of primary and standby resources in the cloud and edge. It needs to solve the problem of streaming processing in high concurrent state. This paper studies the resource scheduling optimization technology adapted to the power cloud edge collaboration, innovatively proposes a multi-data center resource optimization and upgrading method based on the graph data structure, adapts to the multi-center resource optimization and upgrading scenario, uses the RDF resource description framework, TLGM data model to build the multi-data center resource database, uses the global scheduler, the edge scheduler to process the calculation request, uses the data linkage state data model, scheduling rules The probability calculation matrix converts the resource consistency and resource utilization into graph query, and uses the original graph retransmission, subgraph merging technology and efficient load balancing to realize graph query, so as to realize the optimization and upgrading of resources in multiple data centers. This paper innovatively proposes a stream data processing method that is suitable for the cloud edge collaborative multi- data center scenario. It is suitable for the cloud edge collaborative multi-data center stream data processing and analysis scenario. The stream business control and orchestration center construct a serial-parallel collaborative flow analysis process based on the pipeline processing model and parallel processing model. The flow control center control terminal, edge data center, and cloud data center implement flow analysis and scheduling according to the business priority, give full play to the advantages of cloud edge collaborative multi-data center distributed computing to achieve rapid processing and analysis of streaming data.

Tianzhu Tibetan Autonomous County is located at the eastern end of the Hexi Corridor. Huazangsi Town is the seat of the current Tianzhu County. According to the atlas of Chinese dialects, Huazangsi dialect belongs to the Hexi section of Lanyin Mandarin. Its tone system has some distinctive tonal characteristics in the three aspects of single tone, continuous tone sandhi and soft tone. This paper analyzes the characteristics of single tone in Tianzhu Huazang Temple dialect through the research method of experimental phonetics, and summarizes the tonal types, values and areas of single tone in Tianzhu Huazang Temple dialect, providing a teaching basis for the acquisition of Mandarin in this region. At the same time, it explores the development and change of the tone of Tianzhu Huazang Temple dialect under the contact with multi-national languages, analyzes the influence of Mandarin tone on the tone of Tianzhu Huazang Temple dialect, and makes a simple prediction of the future development trend of the tone of Tianzhu Huazang Temple dialect.

With the rapid development of science and technology, the characteristics of comprehensive intersection of modern mechanical disciplines are becoming more and more obvious. Strengthening professional knowledge education, innovative consciousness training, hands-on ability training and comprehensive quality training has become the basic requirement of talent training in mechanical discipline and the core content of contemporary higher education in mechanical discipline. In order to achieve "safe, efficient and high-quality" engineering construction and operation, it is urgent to apply intelligent algorithms to analyze and process engineering big data to provide support for decision-making and planning and design of engineering construction. These demands strongly drive colleges and universities to cultivate talents who understand engineering construction and have intelligent information thinking, which is also the main connotation of engineering professionals training under the background of new engineering. In order to improve students' interest and subjective initiative in learning this course, so that students can deeply understand and master the course knowledge and can learn and use it flexibly in practical engineering, this paper discusses the teaching content and teaching method of mechanical engineering materials course according to the characteristics of safety engineering specialty. The results show that the experimental system has complete functions and good operability and stability.

The application of deep learning technology in target detection algorithm significantly improves the performance of the algorithm. Based on the traditional target detection algorithm, the task of target detection is summarized, including evaluation index, open data set, algorithm framework and the defects of traditional algorithm. Therefore, taking the needs of object detection as the fulcrum, the training goal of research travel talents is clarified in this paper. There are two classification criteria: whether there is an explicit regional suggestion and whether a prior anchor frame is defined. The existing target detection algorithms are classified, and the evolutionary route of each algorithm is reviewed, and the mechanism, advantages, limitations and application scenarios of each method are summarized. The performance of representative target detection algorithms in open data sets is compared and analyzed.

Artificial intelligence is an advanced technology that uses computer intelligence to process data. It can be used for data analysis and decision making. For the application of computer network technology, artificial intelligence and big data can effectively change the previous operation mode, effectively improve the work efficiency of employees. However, there are still many problems to be solved in practical application environment. This paper illustrates the development potential of artificial intelligence in computer network technology through its advantages and applications.

One of the main purposes of music recommendation system is how to recommend the songs that users expect from the massive song data. Most people will use the search function of the software to search for some singers or favorite song categories they have known before. However, the search results do not consider that users are different individuals and have different preferences for songs, which leads to low user satisfaction. Driven by big data, this article proposes a individuation recommendation algorithm for pop music based on deep learning. At present, the music resources on the Internet are extremely rich, and users of various music platforms are facing the troubles of too many kinds of music and difficult to express their emotions while enjoying the leisure time brought by music. By analyzing the music files in the system and the massive user behavior records saved, the user's interest preferences are obtained, and personalized music service content is provided to users. The simulation results show that the individuation recommendation algorithm of pop music in this article is better than the traditional Collaborative Filtering (CF) in recommendation accuracy and user rating.

With the continuous improvement of public aesthetics and more and more music with different changes and styles, the music retrieval system should be more efficient and diversified. However, the traditional music classification system often needs enough perfect music samples at the initial stage of training, and it cannot be effectively adjusted with the addition of various new music samples. At present, most audio music classification algorithms include two stages: feature extraction stage and classification stage. Many musical features can be used to realize this algorithm, including short-time energy and short-time zero-crossing rate in time domain, bandwidth and spectral centroid in frequency domain, and MFCC coefficient based on auditory perception. The task of music style classification is to classify the music into a certain style by processing the data of music signals. Using the music style classification system can help users quickly find music of relevant styles and achieve more effective management of music database. In this paper, Support Vector Machine (SVM) algorithm is used to classify UCI standard data sets. The results show that the learning function with simple structure is adopted for the data set with few training samples. For the dataset with more training samples, the learning function with simple structure will reduce the generalization ability of machine learning.

Three-phase induction motor is widely used in production and life. In the production process of three-phase induction motor, the motor test is very important. The reliability of the test results affects the quality of the motor. In the production and testing process of three-phase induction motor, there are some problems, such as transient overvoltage leading to reduced motor life, inaccurate data due to noise collected, complicated modeling process of performance curve, which can not generate performance curve well. To solve the above problems, we need to design a series of algorithms to improve the reliability of the test. In this paper, the research status of test reliability algorithm of three-phase induction motor is analyzed, which provides reference for the engineering

Aiming at the difficulty in extracting the failure characteristics order of aero-engine intermediate bearing under variable speed conditions, an order tracking algorithm is proposed for a failure diagnosis method of intermediate bearing. This method combines the order of cubic spline interpolation and the ordering algorithm of uniform acceleration to solve the correlative function between the rotating angle of the inside and outside rings of intermediate bearing and time. By the relationship between the rotating angle and time of the inside and outside rings the instantaneous change rate of the rotating angle difference between the inside and outside rings with time is solved. The rotating angle difference sampling frequency is determined. Resample the vibration signal and perform Fourier transform on the resampled signal to extract the failure characteristics order of the intermediate bearing under variable conditions. Verification of the algorithm is shown by the failure simulation signal of the intermediate bearing. The results indicate that the algorithm is very effective for failure characteristics extraction and failure diagnosis of intermediate bearing under variable speed conditions.

This paper studies the static network community detection algorithm, improves the overlapping community detection algorithm LFM based on local expansion, and improves the randomness and redundancy of this algorithm in the process of seed selection and community expansion. Vertex influence and their similarity are used to select seeds, and the redundant steps in community expansion are optimized, and the SCF-LFM algorithm is proposed. The experimental results show that the algorithm has improved accuracy and modularity, and the result of the division has a high degree of community.

In order to solve the problem of low inclusion removal rate in the process of micro-porous argon blowing in tundish, the effects of micro-porous argon blowing position in tundish, porosity of porous brick and bubble size on inclusion removal were studied by numerical simulation. The simulation results show that the porous brick with porosity of 10% are installed downstream of the tundish inlet dam. When the blown bubble size is controlled to 2 μm, the removal rate of 20 μm micro inclusions in tundish reaches the highest 37%, which can effectively improve the purity of molten steel.

Special equipment safety is an important part of the public safety system, which is related to the healthy development of the economy and society and the well-being of people's lives. In recent years, with the rapid growth in the type and number of special equipment, special equipment safety management faces new situations and problems, the traditional inspection and testing model is facing a bottleneck in development, and the Internet of Things technology applied to the inspection and testing system of special equipment, to enhance the level of safety management of equipment has an important role. This paper is based on the current situation of special equipment safety and inspection, the application of Internet of Things technology in the special equipment inspection and testing system to study, in order to enhance the level of special equipment inspection and testing technology, to promote the safety management of special equipment to provide a reference.

As an efficient phase change heat transfer device, vapor chambers are widely used in the field of heat dissipation of electronic elements. The demand for uniform hot plate heat transfer performance is increased with the rapid development of electronic products. In this paper, COMSOL software is used to simulate vapor chamber units. By considering a simplified mathematical model of the construct under certain equivalent assumptions about its working principles, we aim to discuss the relationship between vapor chamber performance and the width of the gas channels inside the wick construct, through analysis of indicative data, including the maximum temperature difference and thermal resistance of the soaking plate.

Contraposing to the problems in image features, grayscale tone performance and artistic effect expression of artistic style rendering technology in simulating handwritten signature to draw character portrait, this paper designs an algorithm to automatically generate portrait by imitating hand painted. Firstly, in order to solve the difference between contour and background in expression, a filtering model based on Sobel operator optimization is proposed to obtain the line texture and contour of the original image as reference, segment the input target image into contour image and background image. Secondly, the background stippling is generated on the basis of the weighted Voronoi algorithm, the stippling is modulated with handwritten signature cells to generate a background image under the definition of signature cells. Among them, after constraining the distribution of cells with the coordinates of the points, the dimension and direction jitter are performed. The size of each cell is changed according to the gray-level integration of Voronoi unit, and the direction adjustment is based on the texture features of the image. Finally, the fusion of contour image and background image is realized by using the local spatial invariance. The experimental results show that compared with the artistic portrait drawn by hand-painted, the image generated by this algorithm not only maintains the original characteristics of the target image, but also achieves similar artistic expression effect. As the product of art digitization, it can be used for miniature printing image, and has unique application value in printing information for anti-counterfeiting.

In order to improve the stiffness and the accuracy of the press structure at work, this paper takes JH31-250 press as the research object and carries out the optimized structural design of the internal support frame in the walls of the press. The press frame model is established in SolidWorks and analyzed numerically in Abaqus for its mechanical properties under work conditions. We found out that the first-order inherent frequency of the press is much higher than the frequency of the excitations at work according to the finite element modal analysis results, so the press action load can be simplified as a static load. Therefore, static numerical analysis of the structure is carried out to assay the stress and the displacement distribution of the structure at work, and the variable density structural topology optimization technique with minimum flexibility as the goal is introduced to find the innovation configuration of the sidewalls. Then, we regularized the design domain of the topology optimization result. After these works, the optimal design of the structure is finally realized. The comparison between the original structure and the optimized structure shows that the low-order frequencies of the structure are changed slightly, but the relative deformation of the optimized structure is largely reduced by 12.4%. Considering the frequencies of various excitation sources during the operation of the press, it can be seen that the optimized structure meets the requirements of use and will not resonate, but the stiffness and accuracy have been greatly improved. The study of this demonstration can be a reference of structural optimization for similar products.

A rapid prediction technique of Sound Transmission Loss (STL) is proposed to investigate the acoustic characteristics of the aerospace vehicle composite panel based on the equivalent method and the hybrid Finite Element method - Boundary Element Method (FE-BEM). The composite panel is equivalent to a single anisotropic plate using the classical laminate theory, and then its STL within a wide frequency band is predicted by the FE-BEM method. To verify the applicability of the equivalent method, modal tests and numerical analyses of the panel are carried out. To verify the correctness of the predicted conclusions, FE-BEM results are compared with FE-SEA results and experimental results. Results show that I) the equivalent method is correct to simulate the natural characteristics of symmetrical laminates, which are consistent with the experimental results; II) it's effective for the panel’s STL prediction via introducing the equivalent structure into the FE-BEM STL prediction model, and III) the FE-BEM method is superior to the FE-SEA method. The proposed method has application in acoustic characteristics analysis and design

To improve the stability accuracy and anti-interference capability of the Fiber Optic Gyroscope (FOG) inertial platform, a Nonlinear Integral Fuzzy Sliding Mode Control (NIFSMC) method that can improve the stability accuracy and anti-interference capability of the inertial platform is proposed. Firstly, a Nonlinear Integral Sliding Mode Control (NISMC) is adopted to rapidly converge the state of the stability loop of the inertial platform to the equilibrium state. Secondly, this article combined fuzzy control and exponential approach law. By designing the fuzzy rules, the parameters of the exponential approach law are adjusted timely. The stability accuracy of the stable loop is improved. Finally, the nonlinear integral fuzzy sliding mode control is verified by simulation experiments. Results show that the NIFSMC method can enhance disturbance rejection ability and improve stability accuracy. Compared with Integral Sliding Mode Control (ISMC) and NISMC, the stability accuracy of NIFSMC is improved by 59.2% and 48.9% respectively.

As the modulus difference exists, road widening is often susceptible to longitudinal cracks at the intersection of the old and new roads, affecting the life of the road and the safety of traffic. For monitoring the differential settlement of the widened road and the creep process of the lower soft soil, a monitoring program is designed and a finite element model is established to figure out the changes in the settlement and earth pressure. Based on the monitoring and numerical simulation, it can be obtained that the settlement difference is comparatively small under 12m mixing pile treatment and the lower soft soil has obvious creep law, which provides a specific case for the usage state of widening road and creep law of soft soil.

In this paper, COMSOL simulation software is utilized to analyze the propagation law of longitudinal wave and the effect of different parameters on ultrasonic axial stress measurement. The results show that there is no reflection phenomenon when the longitudinal wave signal leaves the top of the bolt. When the ultrasonic wave propagates a certain distance, the waveform conversion is generated on the discontinuous structure, and the edge transverse wave is formed after the longitudinal wave, which is accompanied by the entire ultrasonic propagation process. The change of bolt clamping length will modify the slope of the corresponding curve between time and axial stress. By fitting the relationship curve between axial stress and time, the corresponding relationship between axial stress and time is obtained. The experimental error is 3.0%, which proves the accuracy of measuring bolt preload by longitudinal wave method.

This paper proposes an abnormal behavior detection technology of power 5G terminal based on the characteristics of business interaction mode. First, the abnormal behavior detection framework of power 5G terminal is designed. Secondly, considering various security factors of the basic operating environment, a 5G security monitoring indicator system is established around the terminal side security indicators, network side security indicators and API side security indicators. Then, study the automatic learning and construction technology of normal behavior benchmark model, establish a normal benchmark model based on multiple RBM neural networks, study the normal behavior baseline of 5G terminal operation behavior benchmark, the same type of equipment benchmark, specific communication line benchmark and other normal behavior baselines, and build the security baseline indicators of the normal behavior model of 5G terminal. On this basis, abnormal behavior detection technology based on normal benchmark model is studied.

Transmission lines are an important carrier of energy transmission, a key link to ensure the safe operation of large power grids, and an important foundation to support the construction of energy Internet enterprises. In order to adapt to the development situation in the new era and implement the relevant requirements of construction, this paper puts forward the design method of UAV inspection based on transmission lines. The proposed methods include how to create a UAV-based collaborative autonomous inspection mode and improve the autonomous inspection operation and intelligent analysis level of UAVs. The purpose is to implement the construction of unmanned aerial vehicle intelligent inspection operating system and overcome the practical problems of unmanned aerial vehicle autonomous inspection and intelligent recognition of inspection images. Through the construction of collaborative inspection strategy, the advantages of operators of all ages are given full play, the inspection scope is accurate, the inspection efficiency is high, the operation and maintenance operations are accurate, and a new situation of transmission inspection is created.

Electric power is one of the important pillars to safeguard the national economy. Accurate identification of power transmission channel faults is an important means to ensure the normal power supply. In recent years, due to the continuous expansion of the scale of transmission lines, the coverage of the terrain is increasingly complex, resulting in the traditional manual inspection method, has been unable to meet the power supply. Based on this, this paper proposes a power channel intelligent operation and maintenance diagnosis method based on RCNN algorithm. Focus on some objects that are not vertical or parallel and occupy a large proportion across the image. The first challenge was to accurately label such targets. Then the RCNN algorithm model is designed to identify the defects. The experiment shows that it improves the accuracy of recognition to a certain extent.

Due to the rapid development of wireless ad hoc networks, routing protocols in wireless ad hoc networks must have better adaptability to reduce system complexity, save routing overhead and effectively use limited network resources. Currently, the traditional path addressing method using a route-based routing mechanism will cause a large amount of network load and greater delay. To this end, we propose a new method based on Convolutional Neural Network (CNN) to identify, evaluate and guide network features. Using the service characteristics and link status characteristics of the network, comprehensively using the network characteristic matrix as the signal of the network, and adding the measurement of the network performance, the network characteristics can be more completely and accurately identified. The best choice for network performance. Through simulation experiments, it is verified that the method can correctly identify the current network characteristics, make correct predictions about the current network conditions, and make appropriate path selections as needed, thereby effectively reducing network congestion and can effectively solve the network congestion problem.

Distribution network construction is an important work in urban planning and construction, which can provide power supply guarantee for the normal growth of production and living activities, and is the foundation for promoting the rapid and stable growth of urban economy. However, the planning level of the original urban power grid is low, the laying of underground pipelines is not reasonable enough, the quality of power grid equipment can not be guaranteed, and it is prone to operational failures, which leads to great potential safety hazards. As the infrastructure of urban power supply, the power grid's equipment level determines the power supply capacity and affects the security of urban power supply. This paper explores the possible problems in the construction of smart grid, and proposes a power grid planning model based on mobile Internet of Things (IoT) and deep learning, in order to realize the scientific planning and operation of smart power grid as soon as possible. The simulation results show that with the increase of the quantity of experiments, the accuracy of this algorithm is stable at about 94%, and the real-time wavelength tends to be stable, which is 18.64% higher than the traditional algorithm on average. Using this model to transform and brand-new plan the power grid system, research and popularize the automation technology of distribution system, and build a smart grid, it is expected that the smart grid can be more effectively planned and operated.

In the distributed resource access scenario, the growing number of terminal devices brings pressure and test to the wireless communication network. Unauthenticated terminal access to the network will bring huge security risks. This paper focuses on the hardware features of resource aggregation control communication channel and signal source. For the distributed power system with complex environment, this paper proposes a hardware feature extraction technology for resource aggregation-oriented control communication channel and signal source. On the basis of the research on the theory of generic feature extraction of hardware fingerprint, the specific feature extraction methods of several common communication methods in typical distributed resource aggregation and control scenarios are studied. This paper designs a hardware fingerprint implementation scheme in a typical distributed resource aggregation and control scenario. Experiments show that the proposed fingerprint extraction scheme can effectively distinguish different devices according to different characteristics of devices. It provides a terminal access method with high security and low complexity in the distributed edge aggregation control network of power system.

In recent years, with the advancement of aerospace integration and the development of space missions, micro-nano satellites have become a hot research direction in aerospace field. Limited by the performance limitations of single micro-nano satellites, the networking of micro-nano satellite constellation is becoming the main trend. Optimized Link State Routing (OLSR) protocol is widely adopted as the micro-nano satellite networks routing protocol. For the micro- nano satellite constellation networks, frequent changes of the network topology caused by the rapid movement of satellite nodes pose a high challenge to the invulnerability of routing protocols. However, traditional OLSR routing protocol lacks awareness of the satellite node running trajectory and running status, resulting in untimely updates of network topology changes, the invulnerability cannot be guaranteed. To this end, in view of the predictability of the micro-nano satellite constellation orbit, we propose an improved OLSR protocol based on link survival time optimization. By adding the interaction of position and other information in OLSR protocol, link survival time is predicted based on satellite orbit. Extensive simulation experiments show that improved OLSR protocol based on link lifetime optimization has significantly improvement in transmission success rate and invulnerability.

The research of IOT (Internet of Things) technology is the product of the continuous development of the information age. The impact of IOT technology on human production and life is gradually increasing, and IOT encrypted data based on cloud computing platform is also gradually popular. Because ECS is not completely trusted, cloud tenants will face the risk of sensitive data leakage while enjoying convenient data services. In the development history of IOT technology, cloud computing plays an important role. As a large-scale computing method, cloud computing has super data processing capacity, and can complete data scheduling in real time, effectively accelerate data computing speed. It has high development potential. The analysis of IOT encrypted data on the cloud computing platform has become the basic technology for feature extraction and data mining of big data information. The experimental results in this paper can prove that the method in this paper has better key splitting effect and higher key flexibility, which can realize the random splitting of IOT keys, improve the ability of data to resist external attacks, and ensure data security. It is of great practical value to realize the open source IOT encrypted data platform based on the current cloud computing technology, and it also has great significance to promote the development of IOT technology.

In order to process multiple RF signals and realize to output its digital signal. One RF signal analysis device is designed, it can used in radar and other signal analysis fields. This device include a set of software and hardware system, firstly, conditioning the analog RF signal from external input, make its amplitude conform to the requirements of analysis signal, and then convert analog signal to digital signal. This digital signal collected by the FPGA, analysis by DSP again, then, analysis data transmit to FPGA, and finally, FPGA transmit some related data to upper monitor according network, the upper monitor displays the relevant information of the signal through the software developed by the user. The designed device can display, analyze and filter RF signal in time domain. Therefore, the device that analysis signal based on FPGA&DSP architecture has good reliability in the analysis of RF signal.

Various defects of power equipment affect the normal operation of power grid, and serious defects even bring great losses to production and life. Infrared image recognition of power equipment is a necessary prerequisite to realize condition monitoring and fault diagnosis of power equipment under infrared imaging. Because of the imaging characteristics of infrared images, the complexity of background environment and the diversity and difference of power equipment itself, it is difficult to identify infrared images of power equipment. The purpose of this article is to propose a fast and accurate condition monitoring method for power equipment. Based on this, this study proposes an infrared image recognition algorithm of power equipment based on improved Convolutional Neural Network (CNN), which provides technical support for the construction of power equipment condition monitoring system. The simulation results show that the objective function of the improved model can achieve stable payment with fewer iterations, and it is superior to the traditional Support Vector Machine (SVM) algorithm and Ant Colony Optimization (ACO) algorithm in terms of accuracy, recall and running time, thus verifying the effectiveness of the algorithm and the interference to different backgrounds.

In order to make more effective use of distributed power generation and reduce the consumption of reactive power and the value of voltage deviation in the network, a more effective model is established after in-depth analysis of the effects of some new energy power supply equipment and reactive power regulation equipment on the active distribution network. Because the simulation is neither linear nor non-convex, this paper intends to use second-order cone programming to optimize the reactive power of the distribution network and intends to use the second-order cone optimization method in convex optimization to solve the model proposed in this paper. The effectiveness of the proposed algorithm in solving reactive power optimization is verified by the analysis and calculation of an IEEE33-bus distribution system example.

In the era of big data, artificial intelligence is widely used in computer network technology, which makes the functional and structural layers of computer network system comprehensively optimized. The use of natural language generation, speech recognition and virtual assistants make people's lives easier and faster. At the same time, the application of big data artificial intelligence has effectively strengthened the construction of computer network security system, and realized the high-end, intelligent and automatic operation of computer network. In this paper, the concepts of computer network technology and artificial intelligence technology are described in detail. And summarizes the advantages of using artificial intelligence technology in computer network technology and its specific application in computer network technology. Then it makes a detailed analysis and summary of its application strategy in computer network technology. In order to accelerate the development of artificial intelligence technology at the same time to improve the overall effect of computer network technology, so that it further towards the direction of big data development.

The microstructure formation and structural regulation of Mg-Cd ferrites (Cd_0.15Mg_0.85Fe₂O₄) with 5% Bi₂O₃ addition at different sintering temperatures (880, 900, 920, and 940 °C) werer invsetigated for application in VHF antennas. The changes including microstructure, phase composition, magnetic and dielectric properties were mainly characterized. It was found that increasing temperature resulted in superior microstructure with accrescent grain size and denser arrangement. XRD patterns revealed normal spinel phase and a secondary phase of Bi compound. Meanwhile, a reinforcement of the magnetization including increased saturation magnetization (from 37.8 to 46.6 emu/g) and dropped coercivity (from 98.2 to 65.9 Oe) were obtained through hysteresis loops. A slight increment of μ' and ε' were measured over a broad frequency range in VHF bands. Therefore, excellent magneto-dielectric properties of the proposed ferrites were obtained. In addition, low magnetic losses (tan δμ ~10 ^-2) and dielectric losses (tan δμ ~10 ^-3) indicating low power loss in operation. It is foreseeable that these properties of the proposed ferrites would behave well in miniaturized VHF antennas.

In a high-precision fiber optic platform system, temperature changes will have a significant impact on the start-up speed and actual application accuracy. In order to effectively improve the navigation accuracy of the system, the article analyzes the temperature characteristics of the system-level calibration error parameters and establishes a temperature error model. Finally, a temperature compensation test and a platform model calibration test are designed to verify the compensation effect. The test results show that the established temperature error compensation model is accurate and effective. The stability of error parameters has increased by an order of magnitude. The method improves the performance of the fiber optic platform at the start-up stage, and has a small amount of calculation, which has high engineering application value.

Aiming at the problems of low integration, inflexible performance reconstruction and poor stealth performance of traditional metal antenna systems, a reconfigurable antenna is designed by using Si/Ge/Si hetero-SPiN diodes. When the forward voltage is applied on the SPiN diode, the carrier concentration in the intrinsic region is high, and solid-state plasma area is formed, which has metallic characteristics. When the diode is off, it is equivalent to a dielectric and can’t respond to external electromagnetic waves. The generation and annihilation of solid-state plasma is controlled by tuning the applied voltage of different parts of the hetero-SPiN diodes, which realize reconfigurability of antenna performance.

Today, the world economy is in the stage of rapid development, followed by the rising quantity, the decreasing moisture content, and the increasing heat value of Municipal Solid Waste (MSW). Since waste incineration has the advantages of excellent volume reduction effect, high resource utilization efficiency, and less secondary pollution to the environment, waste incineration disposal technology has received extensive attention worldwide. With the continuous increase of the urban population, the amount of sewage sludge is growing year by year. The process of the sludge incineration system is highly complicated, and the cost of construction and operation is high. The waste incinerator can process not only MSW but also sewage sludge. Therefore, the co-incineration of sewage sludge with MSW in the existing waste incinerator is a safe and effective way to dispose of sewage sludge. In this paper, the simulation results of Computational Fluid Dynamics (CFD) are analyzed for medium-scale and large-scale waste incinerators, characterized by high moisture content, high ash content, low heat value of sewage sludge, and incomplete combustion and high pollutant emissions in waste incinerators. With FLIC-Fluent coupled simulation method, the critical information such as temperature field, velocity field, concentration field of principal components, and the pollutant emissions can be predicted. Based on the CFD simulation results, the performance and structure of waste incinerators can be optimized, and the new products can be designed and developed.

In order to solve the problem that suspension insulator in substation cannot effectively implement zero value detection, a terminal tool for suspension insulator resistance detection robot based on wireless remote control is designed and researched, which realizes the autonomous walking of the tool on insulator string in substation and the measurement of insulator resistance. The mechanical design, electrical system design, and experimental studies of the terminal tool for insulator resistance detection robots are presented emphatically. The test results show that the designed terminal insulator detection tool meets the requirements for suspension insulator detection in substations.

In view of the tracking requirements of the intelligent car in the driving process, the control circuit with the STC89C52 MCU as the core is designed. The modular design scheme is adopted, and the color code sensor, metal detection sensor, ultrasonic sensor and Hall sensor are used to form different detection circuits to detect the car's track, detect embedded metal chips, avoid obstacles, measure speed and other problems, The circuit is analyzed theoretically and tested practically. The results show that the intelligent car has good identification and detection capabilities, and has the characteristics of positioning accuracy, stable and reliable operation.

This paper attempts to use a software architecture based on observer mode, establish cloud computing data centers by region, register computing resources to the registry, and take the registry as the system coordinator to ensure that all vehicles can always find the best computing resources in the registry. The process is that all cloud computation registers themselves to the register with their connect info and geographical area. All the intelligent connected should get cloud service information from the register and the point of vehicle can estimate the best cloud server. It ensures that all the intelligent connected vehicles can transfer data to best cloud computing data center get best network and computing services. so as to achieve a reasonable allocation of computing resources and ensure the real-time and stability of data processing.

In recent years, with the development of science and technology and the improvement of informatization, computer network management has become an effective means of overall and standardized hospital management. The hospital information system is constantly improving, which plays an important role in improving clinical efficiency, improving patient satisfaction and unifying the appointment management of medical resources in the whole hospital. With the large-scale coverage of mobile phone 4G services and the effective integration of medical industry and information technology, mobile medical services have gained a good development opportunity. Using mobile phone as the carrier of mobile medical services can, to a certain extent, solve the problems existing in traditional medical treatment activities, such as difficulty in registration, long service time, short face-to-face consultation time and uneven distribution of medical resources. After demand analysis and technology learning, this paper finally designs a hospital appointment registration system based on mobile terminal. First, according to the research results at home and abroad, the technologies needed by the server and client are determined. Then, based on the JAVA framework and the development environment of MySQL database system, the system audience is analyzed, the functional requirements are analyzed, and the feasibility of the scheme is analyzed, and the system database is designed and improved. Finally, through many program adjustments and code debugging, the system can run smoothly and and meet the actual needs.

In order to ensure the normal operation of OBD (On Board Diagnostics) system, manufacturers are required to conduct self-examination and supervision of OBD system, which is the performance evaluation of production vehicles. Multi-objective GA (Genetic Algorithm) simulates the process of biological evolution, dealing with a population, and can generate a large number of non-inferior solutions in one optimization process, so it can search the approximate Pareto optimal solution set of multi-objective optimization problems. In this paper, a fault diagnosis model based on multi-objective GA is proposed to diagnose the faults of automobile engines. The main idea is to use advanced multi-objective GA NSGA-II to adjust the parameters of the production car performance. The simulation results show that the fault samples can be completely and accurately identified by the established model, and the identification rate reaches 91.27%. Therefore, the fault diagnosis model based on multi-objective GA proposed in this paper can be used for fault diagnosis of automobile engine.

This paper proposes an abnormal behavior detection technology of power 5G terminal based on the characteristics of business interaction mode. First, the abnormal behavior detection framework of power 5G terminal is designed. Secondly, considering various security factors of the basic operating environment, a 5G security monitoring indicator system is established around the terminal side security indicators, network side security indicators and API side security indicators. Then, study the automatic learning and construction technology of normal behavior benchmark model, establish a normal benchmark model based on multiple RBM neural networks, study the normal behavior baseline of 5G terminal operation behavior benchmark, the same type of equipment benchmark, specific communication line benchmark and other normal behavior baselines, and build the security baseline indicators of the normal behavior model of 5G terminal. On this basis, abnormal behavior detection technology based on normal benchmark model is studied.

With the development of technology in the power field gradually becoming digital and automated, technologies such as cloud computing and big data are gradually applied to sense and control the power grid. However, accurate sensing and measurement of the grid state is the main technical basis for data analysis and automated operation and maintenance. Therefore, this paper proposes a multi-point synchronous phase measurement method based on non-contact micro-sensors to address the shortcomings of existing grid sensors in terms of sensing accuracy and timeliness. The method solves the synchronous phase measurement problem of grid PMUs by deploying non-contact TMR sensors to achieve single-axis current measurement. And based on this, the multi-point synchronous phase measurement method is improved and designed to solve the theoretical accuracy error caused by wire eccentricity. Meanwhile, it is combined with the dynamic phase measurement method to characterize the synchronous phase change and represent the dynamic characteristics of the grid signal more accurately. The simulation results of the experimental platform show that the measurement error size of the multi-point synchronous phase measurement method is within 4%. Moreover, its phase angle measurement error is less than 0.25 degrees, which can realize the synchronous phase measurement of non-contact sensors. It not only reduces the measurement error caused by the environment, but also makes the measurement more convenient and faster, promotes the digital development of the power grid, and provides great convenience for the fault detection and operation and maintenance of power equipment.

Intangible cultural heritage plays an irreplaceable role in the process of national cultural and economic development. The key factor to solve the problem of intangible cultural heritage protection is to solve the problem of "dissemination". Among them, under the background of contemporary culture, multimedia is not only a tool for people to live and learn, but also an important medium for the protection and dissemination of intangible cultural heritage. The application of intelligent media technology in the inheritance and dissemination of intangible cultural heritage will, on the one hand, break through people's cognitive structure of cultural dissemination, and make the inheritance subjects of intangible cultural heritage more diverse, the forms of dissemination more diverse and the aesthetic experience more profound. On the other hand, it will build a brand-new communication form, realize fission communication, and make the communication of intangible cultural heritage faster, more efficient and more accurate. Intelligent media technology can promote the more effective dissemination and higher quality utilization of intangible cultural heritage by establishing a larger cultural communication platform, stimulating wider cultural consumption, and transforming more cultural resources. This paper proposes that one of the innovation directions of intangible cultural heritage protection is multidisciplinary multimedia design application based on new media application, and finally forms multimedia interactive products with optimized user experience.

Big data technology is the new development of information technology, which can realize the rapid integration of information. Compared with the traditional information integration technology, it is more convenient and faster, and is widely used in all walks of life to obtain product and customer information. In the era of big data, precision marketing is based on data, multi-dimensional analysis of enterprise users' behavior and attribute data, and targeted marketing activities. In this paper, the classification model of e-commerce platform precision marketing based on big data algorithm is constructed. Taking common big data classification algorithms as the theoretical basis of the experiment, this paper introduces the historical behavior data of e-commerce websites as training data, and compares the training results of each algorithm, so as to find and analyze problems and draw conclusions. The experiment uses Python language to realize the data algorithm. The results show that the improved Apriori algorithm can effectively improve the computational efficiency of the algorithm, at the same time improve the bias of Apriori algorithm, and improve the accuracy of the algorithm. Big data classification algorithm can be effectively used in precision marketing to solve classification problems, and the model has high accuracy, which can be used in actual precision marketing.

The popularity of a variety of medical equipment provides the feasibility for integrating medical diagnosis resources and improving the home care of patients. In this paper, the construction of remote intelligent nursing system for the elderly based on machine learning technology is studied. Aiming at the problems of insufficient information mining and low prediction accuracy in multi-task time series, this paper combines the supervised and semi-supervised learning methods in machine learning to predict the physiological status of remote health monitoring objects. The input values are clustered by K-means algorithm, and the similar samples are divided into the same cluster. Each cluster is trained and tested by SVM (Support Vector Machine) algorithm to get the predicted output value. The research results show that the proposed algorithm is more accurate, and the predicted value is closer to the measured value, which shows the effectiveness of the proposed method.

The detection of insulators in substation is completed by climbing the tower manually, which is high labor intensity and high risk. In this paper, the robot system of insulator live detection is proposed. The operator controls the robot system remotely on the ground. The conveying device and detection vehicle are used to place and retrieve the insulator detection end tool. The radar identifies and locates the insulator string and end tool. The end tool detects the resistance values of insulators piece by piece on the insulator string. The resistance values are transmitted wirelessly to the hand-held terminal for display and storage. The robot warns insulators with low or zero resistance. The laboratory test shows that the intelligent degree and accuracy of the insulator live detection robot are high.

Based on the combination of high-speed magnetic bearing and high-speed permanent magnet motor, high-speed maglev centrifugal permanent magnet direct drive frequency conversion unit is the future development direction of air conditioning. In order to make permanent magnet motor control more suitable for high-speed magnetic levitation unit performance requirements. The research on sensorless control of high-speed permanent magnet motor in the background of high-speed magnetic levitation centrifuge is carried out in this paper. The sensorless position observation, maximum torque-current ratio control, flux-weakening control, harmonic and shaft extension suppression, and generation mode of permanent magnet motor are studied. Finally, it is verified in the magnetic levitation unit, which can meet the requirements of the unit.

Pinion and rack transmission system is one of the core components of hoist-way construction hoist. In order to master the load characteristics of gears during operation, based on rigid-flexible coupling dynamics, a joint modeling method using Solidworks, ANSYS and ADAMS software platform is constructed. The rigid-flexible coupling model is established. Taking a practical working condition as an example, the simulation is carried out to analyze the change law of pinion stress under the condition that the pinion and rack are engaged correctly and at a certain angle, and the causes of tooth surface wear during the operation of the system are analyzed. The simulation process and results provide reference and basis for the optimization design of pinion and rack teeth.

E-business has developed rapidly in recent years, and the rapid growth of e-business has promoted the progress of logistics industry. At present, e-business transaction has become the main way of people's life, and the data of logistics system has increased linearly with the growth of e-business. With the continuous promotion of the opening-up policy, transborder e-business has also emerged. However, because transborder e-business is an international business activity, there is a big logistics risk in the growth of transborder e-business. Driven by big data, the growth of transborder e-business will be further upgraded, which will bring more logistics business needs. However, cross-border logistics is difficult to meet simultaneously, which makes logistics risks more and more prominent. This paper discusses the key issues of transborder e-business logistics network and establishes a risk assessment model of transborder e-business logistics network based on Artificial Neural Network (ANN). The experimental results show that the assessment accuracy of this method is over 96%, which can effectively assess the logistics risk of transborder e-business and provide technical support for the logistics network construction of transborder e-business enterprises.

According to the infrared absorption characteristics of gas, a multi-component other analyzer is designed using Lambert-Beer absorption law. The most important SNR influence source of the current high-precision infrared gas absorption sensor has been studied, and the light source drive, infrared gas chamber constant temperature control, signal sampling and weak signal processing circuit design have been optimized. A set of NDIR gas detection device has been designed. After the test of low sulfur gas and carbon dioxide gas, the signal noise fluctuation has been reduced from 140 uV to about 50 uV, and the detection limit has reached 0.01 ppm, the technological innovation of NDIR multi-component gas detection has been realized, and the results of this project have successfully achieved mass production.

Differential evolution algorithm-based tracking control for under-actuated USVs was proposed in paper to deal with the problem of path tracking under external interference. Via the USV’s separated mathematical model, the iterative sliding surface design is carried out by using the hyperbolic tangent function, and the driven inputs of the servo of the ship are deduced. Then, the parameters of the designed sliding mode controller are optimized online through Differential Evolution Algorithm (DEA), which enhance the adaptability performance of the controller. The computer simulation results verify the feasibility of the method.

The narrow waterway has always been a water area with a high incidence of ship collision accidents, and the research on the decision-making of collision avoidance system can improve the safety of ships sailing in the narrow waterway. The geometric scale of ship domain model is conservative, which is not very consistent with the actual collision avoidance of ships in narrow waterway. Therefore, via analyzing the existing research results in ship domain model in narrow waterway, a dynamic mathematical model of ship domain model in encounter situation is established with the "quaternary" ship field and the help of China's port design code, in narrow waterway. The MATLAB simulation platform is used to simulate the avoidance situation of two ships in the narrow waterway, and the feasibility of the model is verified.

With the rapid growth of artificial intelligence, especially deep learning, Natural Language Processing (NLP) is playing an increasingly important role in human learning, work and life. The huge amount of information makes it difficult for people to get valuable information from it. How to quickly screen this valuable information is the key to the intelligent question answering system. In order to obtain knowledge information that is more in line with users' expectations, NLP technology has been researched and developed, and related technical products have been successfully integrated into people's lives, among which intelligent question answering system can better meet people's demand for accurate information. This article explains some limitations and defects of current semantic matching technology, puts forward a local optimization algorithm based on Bert, and applies it to the design of intelligent question answering system. The simulation results show that this algorithm is more accurate for text feature recognition, which is 19.85% higher than the contrast algorithm. The system interacts with users through the visual interface, and automatically replies to the questions raised by users, thus achieving the purpose of practical application.

With the rapid growth and widespread popularity of the Internet, people's access to information is diversified and convenient. The existing text representation methods have some problems, such as insufficient semantic information extraction, high dimension of representation model, and high complexity of model construction. In this paper, a semantic matching algorithm of intelligent question answering system based on BERT is proposed, and the semantic similarity of a concept at the next level pointed by the attribute is found under certain attribute matching rules. Finally, the concept similarity method is recursively called to calculate the similarity of each concept, so that the similarity of all concepts at all levels is weighted and integrated to obtain the semantic similarity between ontologies. The simulation results show that after the system is deployed, the ideal effect can be obtained by comparing the accuracy and response time of text selection. This shows that the improved method proposed in this paper can effectively improve the performance of BERT model, and then effectively reduce the model parameters and accelerate the pre-training speed under the same performance.

With the development of economy and the continuous progress of science and technology, people are increasingly pursuing a highly safe and comfortable living environment and intelligent and diversified information services. Smart home system is generally based on old and new houses, and through a series of wireless wired communication technology, network integration technology and automation application technology, a family management system composed of environmental data collection, terminal control of household appliances, security alarm and other related aspects is established, thus providing a highly integrated and comfortable home living space. In line with this trend, we designed DCA-3000 smart home system, which is suitable for the middle and high-end market. This system can realize the functions of home-based security and remote meter reading. However, the Zigbee communication module, a data acquisition sub-node, adopted at that time, has high power consumption and no sleep function, so it is difficult to supply power with batteries for a long time. On the basis of analysis and design, a temperature acquisition subsystem of smart home wireless network based on ZigBee technology is implemented. Through the implementation of the subsystem and relevant experimental tests, the feasibility of the overall scheme design of the smart home system in this paper is verified. The network composition based on ZigBee technology can meet the requirements of the smart home system.

Sleep is a dynamic process, which changes every day, so it is necessary to continuously measure how many nights of sleep, whether in medicine, research or health diagnosis. The essence of sleep monitoring is to monitor physiological signals, and realize sleep staging, sleep quality analysis and disease diagnosis through monitoring physiological signals. Sleep monitoring technology should not only reduce the interference of monitoring equipment to human sleep as much as possible, but also collect enough sleep physiological information for extraction and analysis. In order to realize sleep monitoring, this article proposes an intelligent data acquisition and mining algorithm based on linear regression model, which provides theoretical and technical support for the design of non-invasive sleep monitoring instrument. The simulation results show that after many iterations, the MAE of this algorithm is 21.66% lower than that of ID3 algorithm. Therefore, the application of this algorithm in the design of non-invasive sleep monitoring instrument can realize automatic sleep monitoring, and then realize intelligent control of electrical switches according to user requirements and user sleep state.

More and more scholars put forward to apply deep learning to chat robots and realize the dialogue processing of chat robots through semantic analysis and emotional dialogue characteristics of deep learning. In actual chat, all kinds of sentences will exist, but there are some uncertainties in the logic of sentences and the part of speech of words, and some questions have different meanings of the same word. In order to carry on the dialogue, the robot must have strong semantic disambiguation ability and remember the answered questions. In this article, an intelligent chat robot semantic matching algorithm based on Seq2seq is proposed, and under certain attribute matching rules, the semantic similarity of a concept at the next level pointed by the attribute is found, and the similarity of each concept is calculated, so that the similarity of all concepts at all levels is weighted and integrated to obtain the semantic similarity between ontologies. The experimental results show that the dialogue generation mechanism designed this time has higher recognition rate than the traditional mechanism, and can correctly identify the dialogue content to make accurate answers, which provides technical support for the development of intelligent chat robots.

In recent years, AI (Artificial Intelligence) technology has been applied to the diagnosis of diseases such as lung cancer, skin cancer and diabetic retinopathy, and has shown excellent diagnostic performance. AI technology with deep learning as the core can make full use of massive data, automatically learn the features in the data, and assist doctors in diagnosis accurately and quickly. Gastric cancer has a high degree of malignancy, hidden onset and no specific symptoms in the early stage. Its clinical manifestations are often similar to benign gastric diseases such as gastric ulcer and chronic gastritis, and it is easy to be ignored. In this paper, the gastroscope image recognition model and diagnosis system based on AI technology are developed. In this study, a gastroscope image recognition model based on GCN (Graph Convolutional Networks) is proposed. The GCN multi-label classification module is responsible for learning and representing the physiological and anatomical relationship of the upper digestive tract, and fusing the learning results with the output results of the detection module. Finally, a gastroscope image aided diagnosis system is designed and implemented. The results showed that the sensitivity and PPV of the model were 95.360% and 91.017%, respectively. It is suggested that the model is superior to the traditional method in the detection rate of early cancer lesions.

Nowadays, enterprise intellectual property management has become the focus of management, which is an effective way for countries to integrate into the trend of economic globalization and strengthen their influence. If enterprises want to achieve long-term and stable development, they must do a good job in intellectual property management, and constantly optimize the management system to improve the management level. Based on this, under the background of the great development of social informatization, this paper fully analyzes the characteristics of intellectual property management. On the basis of summarizing relevant technologies, it uses Spring+Hibemate framework technology and MVC mode to design and implement the enterprise intellectual property management system, which provides users with the functions of intellectual property information management, submission, approval, etc., and verifies the stability and reliability of the system through a series of system tests.

With the improvement of message and internet technique, college English education mode is also facing corresponding changes. It is the most important task of discipline research and improvement to actively combine with internet message technique, update and upgrade college English education technique in time, and effectively combine AI with college English education. The era of AI has come, and the improvement of education is inseparable from AI, and the reform of English education is also deeply influenced by it. AI enriches English education resources, changes teachers' education means and stimulates students' enthusiasm for learning. AI, as a new round of education technique reform, injects new vitality into education. By transferring human intelligence to intelligent machines, AI enables intelligent machines to make human-like thinking responses in different application scenarios. This research is based on AI, the design and simulation of autonomous learning platform for constructive English education. According to the research, this means is effective and suitable for being widely used.

Cardiovascular disease is a serious threat to human life and health, with high prevalence, high disability and high mortality. Cardiovascular disease Q&A system is more targeted and reliable than search engines, and users can get more concise and clear information. As a structured data source of the Q&A system, the knowledge graph enhances the flexibility and accuracy of the system. In this paper, knowledge extraction from authoritative medical websites and Baidu Baike using web spiders, knowledge fusion with reference to medical textbooks on cardiovascular diseases, and knowledge storage using Neo4j graph database were used to build the knowledge graph of cardiovascular diseases. On this basis, the Q&A system was built based on template matching and vector parsing methods, using word2vec model, TF-IDF algorithm, Naive Bayes classifier and support vector machine classifier from the perspective of doctors and patients respectively. The experiments showed that the Q&A system can effectively answer most of the questions on cardiovascular diseases.

In artificial intelligence research, prosodic and pitch features are the focus of singing intelligent speech synthesis. Pitch features are usually studied by extracting fundamental frequencies. In the process of folk song singing, many special vocal skills are used, mainly related to the mode of vocal fold vibration. Using singing voice signals and parameters to study singing pitch and vocal types can better obtain the pitch and prosodic characteristics of singing. This experiment mainly takes Ewenki folk songs as the research object, analyzes singing voice signals and parameters, so as to provide reference for singing voice synthesis. The Ewenki are one of China’s ethnic groups with a small population. The Ewenki folk songs are sung in the Ewenki language, which retains many features of ancient folk songs and singing pronunciation characteristics, especially in the vocal vocalization of the singing voice. The article takes the typical traditional folk song of Ewenki, "Molie Girl", as the object of study. By collecting the singing voice and voice signals of a professional and a non-professional singer, extracted the parameters of fundamental frequency, open quotient and speed quotient. After statistical analysis of the parameters, the overall voice characteristics of the professional and non-professional singers during the singing process and the differences in the voice parameters of the trailing part of the voice were obtained. The analysis revealed both professional and non-professional singers were able to express the artistic characteristics of traditional folk songs, with different vocal styles and singing techniques. The experimental voice acoustic parameters and results can be used as a reference for the synthesis of traditional folk songs and as a method for the teaching of traditional folk songs.

Nowadays, with the rapid development and popularization of informatization and intelligence, "cloud computing service" is also born. Among them, "private cloud" is an important part of "cloud computing services". This is also known as the "internal cloud" or "corporate cloud" and plays an important role in the realization of information management within the enterprise. By studying the overall architecture of private cloud security management platform, this paper studies the key technologies of cloud security management platform and the application value of cloud security management platform. It can provide more flexible security solutions for governments and enterprises and provide better security for computer networks.

In order to ensure the safety and efficiency of emergency transportation process for natural disaster environment and promote the construction of emergency transportation support equipment system, the construction method of road emergency transportation support equipment system is put forward by referring to the construction method of disaster site survival support equipment system in high and cold plateau regions one and combining with the development status of emergency transportation support equipment in China. The boundary, status quo, support objects, support tasks and target capabilities of the emergency transportation support equipment system were analyzed, and framework of the equipment system was constructed, and the evaluation indexes of the system were summarized. Results show that “emergency detection, transportation loading and unloading, special transportation and emergency support are important support objects, their support requirements in the process of emergency transportation for natural disaster environment should be fully demonstrated in combination with the support task and scale.

The characteristics of blockchain technology and emergency logistics have many internal points of convergence, which is conducive to promoting the realization of efficient, reliable and intelligent security, improving the construction level of emergency logistics, and assisting epidemic prevention and control and emergency management. Based on this, this paper expounds the definition and characteristics of emergency logistics supply chain, constructs the basic model of emergency logistics supply chain based on blockchain technology on the basis of general blockchain technology architecture, designs the functions of the supply chain system, and puts forward policy suggestions, hoping to provide useful reference for the construction of emergency logistics supply chain.

Maintenance of the oil blast on-load tap changer (hereinafter: OLTC) is an important process in periodic maintenance of the load-ratio voltage transformer. At present, the rapid increase in the number of the load-ratio voltage transformers poses a challenge to traditional maintenance methods. In this paper, an integrated special device for maintenance of the oil blast OLTC is developed, functions of which includes high-pressure, power-driven grinding, visualization and automatic control. The practical application results in the maintenance site verify the validity of the special device. Compared with the traditional maintenance method, the application of the special device shortens the maintenance time, improves current carrying capacity and insulation strength of the oil blast OLTC, which is of great significance to improve power supply reliability and economy of power grid.

When the clearance fit is applied for the positioning of the cylindrical surfaces spline structure of the engine, the misalignment between the two connected rotors is large, and the inner and outer splines are prone to produce internal friction during the rotor working, and the risk of rotor instability is high. By analyzing the relative motion of inner and outer spline, the expression of work by internal friction is derived, as well as instability threshold speed of rotor with spline structures is further analyzed when the internal damping is considered. The Newmark-β method have been used to simulate dynamic characteristics of established single-disc cantilever rotor model and the effects of critical parameters such as rotor damping, friction coefficient, and misalignment about stability of the rotor with coupling sleeve structure are discussed. The results show that the increase in friction coefficient and misalignment will aggravate the instability of the rotor system. When the modal damping ratio of the rotor is less than 1%, the sub-harmonic frequency appears in rotor vibration and the instability frequency is the first-order natural frequency of the rotor. When the modal damping ratio of the rotor increase to 4.5%, the sub-harmonic components disappear in vibration. Therefore, increasing the damping of rotor can restrain instability of rotor system.

In order to study the sliding mode control method of hypersonic aircraft, the six-degree-of-freedom attitude motion model of hypersonic aircraft was established, and the corresponding linearization and decoupling model was obtained by using dynamic inverse method. On this basis, a sliding mode controller based on exponential approach rate is designed, and the controller is reconstructed by introducing the method of instruction parameterization, which greatly reduces the workload of control parameter tuning. In the end, Continuous Action Reinforcement Learning Automata (CARLA) is improved to optimize the new control parameters. Simulation shows that the algorithm can tune out a set of high-quality control parameters in 100 iterations, enabling the control system to quickly and accurately track a given instruction.

At present, the dynamic modeling methods of elastic aircraft mainly include two kinds: mean axis method and quasi-axis method. The two methods have their own advantages and disadvantages, and there are few articles on the comparison of the two methods. In this paper, based on the Lagrangian equation of energy theory, the modeling methods of the mean axis system and the quasi-axis system are derived, and the elastic models obtained from the quasi-coordinate system and the mean axis system are compared and analyzed. The correctness of the hypotheses is verified by numerical calculation according to the hypothetical conditions, which lays a solid foundation for the further study of aeroelastic modeling.

Overhead transmission lines are subjected to long-term vibration loads, which can easily produce cracks and broken strands, affecting the safe operation of the power grid. Steel-core aluminum stranded conductor is commonly used in overhead transmission lines, conductor strand structure is complex, therefore, the need to "nominal" bending stiffness to characterize its bending capacity, that is, the equivalent bending stiffness of overhead transmission lines. When the equivalent bending stiffness of the conductor suspension point and near the exit of the fixture changes due to fatigue, it will amplify the vibration level of the place, which in turn affects the arc sag and air gap of the conductor. The paper for the occurrence of transmission line breeze vibration characteristics, analysis of the equivalent bending stiffness near the suspension point of the tension tower conductor, the establishment of a 3D model of the conductor, using a large sparse solution matrix as a calculation method, the calculation of the equivalent bending stiffness of the conductor under different constraint length,on this basis, further analysis of the number of broken strands andbroken strand displacement on the equivalent stiffness of the impact. The calculation results show that the equivalent bending stiffness of the conductor decreases with the increase of the constraint length, decreases with the increase of the number of broken strands, and shows a trend of first decreasing and then increasing with the increase of the displacement of broken strands.

High-speed (1761r/min~2077r/min) gear shaft is an important transmission component of wind turbines. Its thermal characteristics have an important influence on the transmission power. Therefore, in order to reduce the chance of wind turbines’ failure, study on high-speed axle’s temperature field is proposed at this paper based on transmission principle using heat transfer as well as method of finite element. This assembly model for wind turbines is firstly established in SolidWorks. Then, analysis of high-speed shaft’s temperature field is conducted with ADAMS simulation data that is input into Ansys workbench. The finite element model is established. Finally, the stress caused by temperature is superimposed to the mechanical stress. Simulation results show that with the increase of misalignment in high-speed shaft, the circumferential force on the gear shaft is also increasing. Thus, the temperature is going up by inches. Deformation also increases. Analysis of gear shaft’s temperature field has laid a foundation for the study of tooth surface bonding, gear bearing capacity, tooth modification and so on, it can also provide a basis for the future temperature prediction and fatigue life analysis.

Because of the mechanical inertia of the mechanical governor, the sensitivity of speed regulation is reduced, which makes the diesel engine speed change range is large in the working process, resulting in poor stability of the engine, reduced economy and power, affecting the overall efficiency of the engine. To solve the above problems, we designed solenoid differential transformer electronic governor, its principle is magneto electric speed sensor to collect the speed of diesel engine. The signal is amplified by the converter and PID adjustment operation, and the driving current is output by the solenoid differential transformer. Under the action of magnetic field force, the keeper moves in a straight line. So, by controlling the current size of differential transformer to adjust the position of the fuel supply rod of the fuel injection pump, fuel supply can change and make diesel engine speed ratio stable.

The principle of the steering system of the tracked vehicle is introduced, and the mathematical model of the steering system of the tracked vehicle is derived. The dynamic model and flow regulation model of the tracked vehicle are built by Matlab/Simulink. The dynamic simulation and comparative analysis of the high-speed steering process of the tracked vehicle in two different scenarios of plane and slope are carried out, which improves the steering performance of the tracked vehicle and provides guidance for the high-speed steering of the tracked vehicle.

Aiming at the tracked vehicle with coupled pump-motor hydraulic steering system, the vehicle system coupling model based on MATLAB/Simulink was established under the braking, steering and straight driving scenario. The mathematical model of hydraulic steering regulation system with variable pump and quantitative motor is established. After analyzing the kinematics and dynamics of the tracked vehicle, the vehicle system coupling model under the above working scenario was established. The main performance parameters of the vehicle, such as track speed, steering radius, steering driving force and steering high pressure, are studied under the conditions of different variable pump swash plate inclination angle (different pump displacement), different vehicle braking force and different engine speed. The stability of the combined pump-motor hydraulic steering system is evaluated, and the correctness of the established model is verified, which has certain theoretical guiding significance for the research of the steering control strategy of tracked vehicles.

The hydraulic-mechanical differential steering enables the tracked vehicle to traverse the complex terrain environment. Therefore, this paper analyzes the principle of the mechanical differential steering of the tracked vehicle, establishes the mathematical model of the variable pump quantitative motor, then the dynamic model of the slope center steering is analyzed. The variation law of required traction force and azimuth angle under different braking forces and slope inclinations is analyzed to determine whether stable steering can be achieved based on MATLAB/Simulink.

In order to study the safety of external thermal insulation system, the typical rock wool thermal insulation system was analyzed as an example. This paper introduces the basic structure and component materials of the rock wool insulation system, and according to the relevant standards, the theoretical calculation of the force of the insulation system under the dead weight, wind load and earthquake action corresponding to different bond rates is carried out, the safety of rock wool insulation system is analyzed. According to the analysis results, it provides the corresponding technical reference for the development and application of rock wool insulation system, and put forward safety quality control measures.

In this paper, a sheathless separation device is designed to solve the problem of insufficient controllability and low throughout due to the introduction of sheath flow in conventional Standing Surface Acoustic Wave (SSAW) microfluidics. The spatially reciprocating microchannel structure is matched with the SSAW pressure field to achieve focusing before separation of particles. The simulation of displacement field on the substrate surface and acoustic pressure field in the microchannel verifies the reliability of design. Meanwhile, the effects of average inlet flow rate and acoustic pressure amplitude on the particle motion trajectory are analyzed in detail, and the results show that these two parameters can be adjusted to separate particles of different sizes, and further give a reasonable parameter interval to complete particle separation. The above results provide a reference and guidance for the design and experiment of sheathless separation devices.

Finite element simulation model for cable brackets commonly used in commercial aircraft was developed to investigate the effect of the stamping depth and stiffeners distribution on the load-bearing capacity of 2A12 aluminum alloy brackets, as well as their failure locations. The results showed that: the stiffener can increase the bearing capacity of the bracket, the greater the stamping depth of the stiffeners, the greater the bearing capacity of the bracket; the stiffeners at both ends of the bending axis of the bracket increased the bearing capacity of the bracket the least, 9%. The stiffeners dividing the bracket bending axis into equal lengths increased the bearing capacity of the bracket the most, 30%. The junction of the horizontal plane of the bracket and the curved surface had the greatest equivalent plastic strain and was most vulnerable to plastic deformation.

Aiming at the problems of lack of intelligent fire extinguishing means, insufficient timeliness of manual fire extinguishing, and difficult security guarantee in the existing valve hall, an intelligent fire extinguishing robot system for the converter station valve hall was developed this paper. The overall structure, robot body, water supply system, navigation system and software architecture of the system were introduced. Water spray test was conducted, and the robot could switch between columnar spray and fog spray. Finally, the robot has been applied in the valve hall for one year. The results showed that the robot has a high degree of intelligence, and can extinguish the fire automatically, accurately and quickly. The robot is suitable for firefighting of equipment in valve hall of converter station and provides a new technical means and method.