基于PSO-NGM模型的电力电子电路故障预测方法

Research on Fault Prediction Method of Power Electronic Circuits Based on Particle Swarm Optimization Non-homogenous Grey Model

针对现有电力电子电路故障预测技术的不足,提出了将电路特征性能参数和粒子群非齐次灰色模型PSO-NGM(particle swarm optimization non-homogenous grey model)模型结合,对电力电子电路进行故障预测。以Buck-Boost电路为例,选择电路输出电压作为监测信号,提取输出电压平均值和纹波值作为电路特征性能参数,并利用PSO-NGM预测模型实现故障预测。实验结果表明,利用PSO-NGM对电路输出平均电压和输出纹波电压的预测相对误差很小,能够跟踪故障特征性能参数的变化趋势,有效实现电力电子电路故障预测。

Aiming at the issue existing in the fault prediction technique of power electronic circuits, this paper proposes that the char- acteristic parameter data is used with the particle swarm optimization non-homogenous grey model（PSO-NGM） to predict the power electronic circuits failure. The Buck-Boost converter circuit is taken as an example to predict its failure,The output voltage is selected as monitoring signal and the average voltage and ripple voltage are extracted as characteristic parameters, then the PSO-NGM algorithm is used to predict Buck-Boost converter circuit. The experimental results show that using the PSO-NGM algorithm to predict the average voltage and ripple voltage, its error is smaller. The new method can be used to trace the characteristic parameter trend and predic the failure of power electronic circuits effectively.