Volume 40, Number 3, June 2022
|Page(s)||645 - 650|
|Published online||19 September 2022|
Power converter fault classification method based on multi-feature selection algorithm
Science and Technology on Avionics Integration Laboratory, China Institute of Aeronautical Radio Electronics, Shanghai 200233, China
2 School of Electronics and Information, Northwestern Polytechnical University, Xi’an 710072, China
In the process of fault diagnosis for the core components of the integrated modular avionics power conversion module, selecting appropriate features can effectively improve the efficiency and classification accuracy of the model, and greatly reduce the computational complexity of the learning algorithm. This paper first designs a typical Sepic structure DC-DC converter model to simulate the typical fault types of the DC-DC converter; secondly, the corresponding original data is obtained through simulation; after data preprocessing, feature extraction and using multiple feature selection fusion algorithm, BP neural network method is used finally for fault diagnosis analysis of DC-DC converter. The simulation verifies the effectiveness of the above method.
对综合模块化航电电源转换模块的核心部件进行故障诊断的过程中, 选择合适的特征能够有效提高模型的效率和分类准确率, 极大地降低学习算法的计算复杂度。设计了典型的Sepic结构DC-DC变换器模型, 对DC-DC变换器的典型故障类型进行故障模拟; 通过仿真获取相应的原始数据, 采用数据进行预处理、特征提取与多特征选择融合; 利用BP神经网络方法对DC-DC变换器进行故障诊断分析, 仿真验证了该方法的有效性。
Key words: feature selection / BP neural network / fault diagnosis / power converter / Sepic structure
关键字 : 特征选择 / BP神经网络 / 故障诊断 / 功率变换器 / Sepic结构
© 2022 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.