The inverse problem of electrical impedance tomography (EIT) is often a highly nonlinear ill-posed problem of image reconstruction and the imaging feature prior information is limited. Since traditional iterative algorithms are not able to process the blurred EIT images, a deep learning algorithm is proposed in this study that can autonomously learning important imaging characteristics by some representative samples to improve the image reconstruction quality. First, a PSPNet network based on MobileNet are preliminary proposed to the EIT image post-processing, some simulation experiments based on the EIDORS platform were designed to illustrate the effectiveness and universality of this algorithms. Then we used the NOSER algorithm to reconstruct images of different shapes of targets, including circle, triangle, square, and any combination of two shape targets. For these resulting images, we used PSPNet for postprocessing. Furthermore, we designed a tank test with the EIT systems. Through detection of single target and two target positions in three different positions, the proposed deep learning PSPNet algorithm can improve the Size Error (SE) index by 9.02% on average, and reduced the Position Error(PE) index by 1.59% compared with the traditional NOSER algorithm. It can be concluded that the deep learning scheme PSPNet can effectively recognize the object positions and recover the sharp contour of targets very well.
A large flexible sensor based on electrical impedance tomography (EIT) is limited in detection area and resolution. It can be improved by introducing center electrodes. However, the number, the position and the new drive patterns will have a great impact on the performance of the sensors with large area. Based on the typical 16-electrode flexible sensor, one, two and four center electrodes are introduced in the design to improve the sensors’ resolution, and the new drive patterns for current injection and voltage measurement are proposed. In evaluating the performance of flexible sensors for the different drive patterns, the detecting resolution and position error are obtained for different numbers of center electrodes. Simulation results show that the flexible sensor with two center electrodes has a satisfactory detection accuracy. And one of the center electrodes is used for current injection and the other is used for voltage measurement. Mainly considering the resolution and position error, the particle swarm optimization (PSO) algorithm is used to optimize the position of the center electrodes. The simulation results show that when the center electrodes are positioned at 0.24 from the center, the flexible sensor has a better performance for multi-objective recognition. Compared with the typical 16-electrode flexible sensor, the detection position error can be reduced by 85.1%. This study provides a new method for finding the suitable number and position of the center electrodes for the design of large EIT based flexible sensors.
A modified magneto-rheological rubber (MRR) which resistance can be reduced to 1KΩ was fabricated by dispersed multi-walled carbon nanotubes (MWNTs) and carbonyl iron particles (CIPs) into polydimethylsiloxane (PDMS) matrix. Based on the excellent piezo-resistive properties of this modified MRR, a MR bearing with self-sensing characteristics was proposed and systematic researched. In order to study the self-sensing characteristics of the MR bearing based on modified MRR under load and magnetic fields, the structure and working conditions of the bearing were simulated and compared. The optimal structure size of the bearing was selected and used for the build-up of experiment test system. The results showed that, under the action of preload condition by bearing, due to piezo-resistance and magnetoresistance behavior of the modified MRR, the electrical resistance of it also can be changed over 28%. It suggests that, in the MR semi-active vibration isolation system, the modified MRBs are potentially capable of being used as an actuator and sensor in the same time.
In this article, we propose a 3D FEM model to analyze the anisotropic thermal behavior of oriented GNPs/polymer composites. The model considers the effect of microstructural characteristics such as the shape and aspect ratio of filler, interfacial thermal resistance, volume fraction, dispersion state and orientation of GNPs dispersion to simulate a steady state heat flow in the composite. In the simulation model, the composite is treated as a cubic unit cell and GNPs are modeled as oblate spheroid with a long axis diameter of 1μm and the aspect ratio is in the range of 0.01 to 0.05. The oblate spheroids align in the cubic to represent the microstructure of the oriented GNPs/polymer composites. By varying the number of the GNPs in the representative unit cell, the volume fractions of fillers are determined. The simulation results show that the parallel thermal conductivity (//) increase in a nonlinear trend and the perpendicular thermal conductivity (⊥) is slightly linear increasing. Compared with experimental results, the trends predicted by the finite element models are consistent.
We perform a theoretical simulation to investigate how the magnetic field induces the interaction of ferromagnetic particles inside MRE, and consequently how the surface microstructures of MRE are changed. Firstly, we propose a mesoscopic model of the MRE by assuming that particles have the same radius, and they are randomly distributed in the matrix. Both particles and matrix are considered as linear elastic materials. We apply Monte-Carlo method to produce the initial surface microstructure of the MRE before considering the magnetic field effect. Further, we use sequential decoupling FEM method to solve the magneto-mechanical coupling problem. The relationships of surface roughness of MREs with the volume fraction of ferromagnetic particles, the magnetic field strength, the initial surface microstructure, and the matrix modulus have been numerically investigated.
This paper proposes an extended Preisach model in which the combination of magnetic field and mechanical stress are treated as an effective field and Preisach diagram is used to qualitatively analyze magnetomechanical behaviors for any kinds of magnetization history. Magnetomechanical behaviors under a certain kind of magnetization history are discussed. The corresponding experiments are carried out. The results show that it is helpful in forecasting the trend of magnetomechanical behavior which gives a reasonable understanding of the various magnetomechanical behaviors that appeared to be contradictory. Moreover, it is useful in improving the stress measurement accuracy by determining the magnetic condition.
As the weakest part in the bridge system, traditional bridge bearing is incapable of isolating the impact load such as earthquake. A magneto-rheological elastomeric bearing (MRB) with adjustable stiffness and damping parameters is designed, tested and modeled. The developed Bouc-Wen model is adopted to represent the constitutive relation and force-displacement behavior of an MRB. Then, the lead rubber bearing (LRB), passive MRB and controllable MRB are modeled by finite element method (FEM). Furthermore, two typical seismic waves are adopted as inputs for the isolation system of bridge seismic response. The experiments are carried out to investigate the different response along the bridge with on-off controlled MRBs. The results show that the isolating performance of MRB is similar to that of traditional LRB, which ensures the fail-safe capability of bridge with MRBs under seismic excitation. In addition, the controllable bridge with MRBs demonstrated the advantage of isolating capacity and energy dissipation, because it restrains the acceleration peak of bridge beam by 33.3%, and the displacement of bearing decrease by 34.1%. The shear force of the pier top is also alleviated.
A magneto-rheological bearing (MRB) is proposed to improve the vibration isolation performance of a floating slab track
system. However, it’s difficult to carry out the test for the full-scale track vibration isolation system in the laboratory. In
this paper, the research is based on scale analysis of the floating slab track system, from the point view of the
dimensionless of the dynamic characteristics of physical quantity, to establish a small scale test bench system for the
MRBs. A small scale MRB with squeeze mode using magneto-rheological grease is designed and its performance is
tested. The major parameters of a small scale test bench are obtained according to the similarity theory. The force
transmissibility ratio and the relative acceleration transmissibility ratio are selected as evaluation index of system
similarity. Dynamics of these two similarity systems are calculated by MATLAB experiment. Simulation results show
that the dynamics of the prototype and scale models have good similarity. Further, a test bench is built according to the
small-scale model parameter analysis. The experiment shows that the bench testing results are consistency with that of
theoretical model in evaluating the vibration force and acceleration. Therefore, the small-scale study of
magneto-rheological track vibration isolation system based on similarity theory reveals the isolation performance of a
real slab track prototype system.
The urgent train braking could bring structural response menace to the bridge under passive control. Based on the analysis of breaking dynamics of a train-bridge vibration system, a magnetorheological elastomeric bearing (MRB) whose mechanical parameters are adjustable is designed, tested and modeled. A finite element method (FEM) is carried out to model and optimize a full scale vibration isolation system for railway bridge based on MRB. According to the model above, we also consider the effect of different braking stop positions on the vibration isolation system and classify the bridge longitudinal vibration characteristics into several cases. Because the train-bridge vibration isolation system has multiple vibration states and strongly coupling with nonlinear characteristics, a human-simulated intelligent control (HSIC) algorithm for isolating the bridge vibration under the impact of train braking is proposed, in which the peak shear force of pier top, the displacement of beam and the acceleration of beam are chosen as control goals. The simulation of longitudinal vibration control system under the condition of train braking is achieved by MATLAB. The results indicate that different braking stop positions significantly affect the vibration isolation system and the structural response is the most drastic when the train stops at the third cross-span. With the proposed HSIC smart isolation system, the displacement of bridge beam and peak shear force of pier top is reduced by 53.8% and 34.4%, respectively. Moreover, the acceleration of bridge beam is effectively controlled within limited range.
We propose a method to analyze and design a laminated MRE bearing, in which the optimal parameters of materials and mechanical structure of the MRE bearing are determined. Based on the multi-scale and magneto-mechanical coupling theories, we establish a comprehensive model for the MRE bearing considering the influence of particle volume fraction, particle distribution, and thickness of MRE laminated layers on its mechanical performance. Within the micro-scale analysis, the representative volume unit (RVU) is used to address the effect of particle volume fraction and distribution on mechanical and magnetic properties of MRE itself. Within the macro-scale analysis, we build both mechanical and magnetic models for the laminated MRE bearing. Based on the theoretical analysis, a laminated MRE bearing with four-layer MRE is designed and fabricated. The performance of the MRE bearing has been tested by using MTS test bench. The results are compared with that of model analysis. It demonstrates that the proposed method can be a useful tool in the development of laminated-MRE bearings for practical applications.
The goal of this study is to investigate dynamic properties and the total energy change of a bio-inspired spider web. To better understand performance, the effects of preload, radial and spiral string stiffness and damping ratio on the natural frequency and total energy of the web are theoretically examined. Different types of web materials and configurations, such as damaged webs are investigated. It is demonstrated that the pretension, stiffness and damping ratio of the web’s strings can significantly affect the natural frequency and total energy of the full and damaged webs. In addition, it is shown that by increasing the pretension in the radial strings one can compensate for the damaged strings and increase the capability of the damaged web to reach that of the full web.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.