支持向量机
断层(地质)
整流器(神经网络)
算法
计算机科学
核(代数)
特征向量
电压
功率(物理)
工程类
人工神经网络
人工智能
数学
地质学
组合数学
循环神经网络
量子力学
物理
电气工程
地震学
随机神经网络
作者
Xiaosong Wang,Dazhi Wang,Yongliang Ni,Keling Song,Yufei Qi
标识
DOI:10.1109/ccdc58219.2023.10326663
摘要
With the rapid development of power electronic circuits, power electronic devices are becoming more and more complex, with many fault types and low diagnostic accuracy. This paper proposes a solution based on variational mode decomposition and sparrow search algorithm to optimize support vector machines. Firstly, Simulink is used to establish the simulation model of three-phase bridge type fully controlled rectifier circuit, and the output voltage signal is decomposed by using variational mode decomposition to calculate and extract fault eigenvalue, and the fault data is normalized; Then the sparrow search algorithm is used to optimize the support vector machine, and the penalty factors and kernel parameters of the optimal estimation are searched out, and the obtained data are trained and diagnosed; Finally, the method is compared with the support vector machine without optimization. The final simulation results show that the method is superior to the unoptimized support vector machine in training accuracy and diagnosis effect.
科研通智能强力驱动
Strongly Powered by AbleSci AI