脉冲宽度调制
电子工程
计算机科学
数字控制
粒子群优化
电容器
整流器(神经网络)
工程类
电力电子
控制理论(社会学)
电气工程
控制(管理)
电压
算法
机器学习
循环神经网络
人工智能
随机神经网络
人工神经网络
作者
Sihui Zhang,Wensheng Song,Hu Cao,Tao Tang,Yuchao Zou
标识
DOI:10.1109/tpel.2023.3307415
摘要
A digital simulation and health monitoring method for single-phase two-level pulsewidth modulation (PWM) rectifiers based on digital twin technology is proposed in this article. First, the closed-loop control and the main circuit are discretized to develop the digital model of the PWM rectifier. Then, the external characteristics of the digital model are compared with the data sampled from the actual physical circuit, and the particle swarm optimization algorithm is used to iteratively optimize the key parameters. Ultimately, the offline digital twin model is developed, which can realize the health status monitoring of insulated-gate-bipolar-transistor power devices, the ac-side inductor, and the dc-link capacitor in the rectifier. Finally, an experimental prototype is built to test the proposed digital twin model and health monitoring method under different load conditions. A comprehensive statistical comparison of the digital twin model and experiment results has verified that the proposed method can achieve health monitoring within the acceptable range, even if under different initial values of parameters. Thus, this article provides a feasible and noninvasive solution for the digital simulation modeling and health monitoring of single-phase PWM rectifiers, without additional sensors and hardware circuits.
科研通智能强力驱动
Strongly Powered by AbleSci AI