Paper
25 September 2023 A new distribution network alert state adjustable potential assessment method based on improved isolated forest algorithm
Weiliang Liu, Min Zhou, Xin Zhu
Author Affiliations +
Abstract
The current conventional distribution network alert state adjustable potential assessment method mainly obtains the assessment matrix by constructing the affiliation function to achieve the adjustable potential assessment, which leads to poor assessment accuracy due to the lack of analysis of line loss abnormal state. In this regard, a new distribution network alert state adjustable potential assessment method based on the improved isolated forest algorithm is proposed. Combined with the isolated forest algorithm, a random hyperplane cutting space is constructed to calculate the line loss anomaly fraction, and the value is used as one of the evaluation indexes to construct a potential evaluation index system, and the actual values of different indexes are calculated to realize the adjustable potential evaluation. In the experiments, the proposed method is verified for the evaluation accuracy. The experimental results show that the error value between the evaluation score and the actual result is small when the proposed method is used to evaluate the adjustable potential of the alert state of the distribution network, which has a more desirable evaluation accuracy.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Weiliang Liu, Min Zhou, and Xin Zhu "A new distribution network alert state adjustable potential assessment method based on improved isolated forest algorithm", Proc. SPIE 12788, Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023), 127885W (25 September 2023); https://doi.org/10.1117/12.3004354
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Power supplies

Fuzzy logic

Data modeling

Compliance

Integration

Power grids

Reliability

Back to Top