高压直流变压器中直流电源故障诊断方法的改进研究

摘要

The diagnosis of the DC power supply fault of the high voltage DC transformer in the traditional method is based on experts' experience and historical data,which has long time,low precision and low reliability.This paper presents an improved diagnosis method which is based on combination of the particle swarm (PSO) and BP neural network,and analyzes the fault type of the DC power supply and its causes to form the training sample.And the training sample is input into the BP neural network model for learning and simulation calculation,and the PSO iterative algorithm is used to track the individual extreme and global extreme,and the velocity and position of the particle are updated to find the optimal solution and realize the fault diagnosis.The experimental results show that the proposed method can effectively complete the fault diagnosis of DC power supply,and has a great advantage in the time,accuracy,and error control methods.%传统方法对高压直流变压器中直流电源故障的诊断,多基于专家经验和历史数据,诊断时间长、精度低、可靠性差.提出基于粒子群(PSO)与BP神经网络相结合的高压直流变压器中直流电源故障诊断改进方法,分析直流电源的故障类型及其产生原因,形成训练样本;将训练样本输入BP神经网络模型,进行学习和模拟运算,采用PSO迭代算法追踪每一种故障类型的个体极值和全局极值,更新粒子的速度和位置,寻求最优解,实现高压直流变压器直流电源的故障诊断.实验证明,该方法能够有效地完成直流电源故障诊断,在诊断时间、精度、及误差控制等方面具有较大优势.