DEFECT PREDICTION METHOD FOR POWER TRANSMISSION EQUIPMENT BASED ON GREY WOLF OPTIMIZATION ALGORITHM AND LSTM-HNA

作     者：Yang, Zhennan Zhang, Jie Zhang, Ke Sun, Youqiang Huang, Wenli Zou, Le 

作者机构：School of Artificial Intelligence and Big Data Hefei University Hefei230601 China Institute of Intelligent Machine Hefei Institute of Physical Science Chinese Academy of Sciences Hefei230031 China Anhui NARI Jiyuan Power Grid Technology Co. Ltd Hefei230088 China 

出 版 物：《UPB Scientific Bulletin, Series C: Electrical Engineering and Computer Science》 (UPB Sci. Bull. Ser. C Electr. Eng. Comput. Sci.)

年 卷 期：2024年第2024卷第4期

页      面：125-140页

核心收录：

基　　金：The authors gratefully acknowledge the support of the \u201CAnhui Province Science and Technology Major Special Project (202203a05020023)\u201D 

主　　题：Long short term memory 

摘      要：Predicting defects in power transmission equipment is crucial for ensuring the stable operation of the power grid. However, existing methods suffer from shortcomings such as ignoring temporal features, susceptibility to irrelevant feature interference, and lack of hyperparameter optimization. To address this, this paper proposes a novel method for defect prediction in power transmission equipment that combines the Grey Wolf Optimization (GWO) algorithm with Long Short-Term Memory networks and an attention mechanism. This method utilizes LSTM networks to extract temporal feature information, innovatively designs a Hidden-layer Neuron Attention module (HNA) to reduce the impact of irrelevant features, incorporates the Grey Wolf Optimization algorithm for automatic hyperparameter tuning, and proposes a joint training strategy for GWO and LSTM-HNA to improve efficiency. Extensive experiments validate the effectiveness of the proposed method, achieving a high accuracy of 97.37% in defect prediction tasks for power transmission equipment. Precision, recall, F1 score, and other metrics outperform other methods, providing a new approach to enhancing the monitoring and fault prevention capabilities of power transmission equipment. © 2024, Politechnica University of Bucharest. All rights reserved.