基于IPSO-BP的地铁牵引逆变系统可靠性预测

Reliability Prediction of Subway Traction Inverter System Based on IPSO-BP

为了提高对地铁列车牵引逆变系统可靠性预测的准确性,提出了一种基于改进粒子群算法优化BP神经网络(IPSO-BP)的可靠性预测模型。引入平均相对误差指标作为可靠性预测模型的评价指标,通过收集到的实际数据对所构建的可靠性预测模型、BP神经网络可靠性预测模型及PSO-BP神经网络可靠性预测模型进行对比验证。仿真结果表明IPSO-BP神经网络可靠性预测模型平均相对误差比BP神经网络、PSO-BP神经网络可靠性预测模型平均相对误差分别降低了约17.5%和10%,具有较好的预测精度,明显地提高了地铁列车牵引逆变系统的可靠性预测的准确性。同时对制定预防性维修周期具有重要的参考价值。

In order to improve the accuracy of the reliability prediction of the subway train traction inverter system, a reliability prediction model based on improved particle swarm optimization algorithm BP neural network(IPSO-BP) is proposed. The average relative error index was introduced as the evaluation index of the reliability prediction model, and the reliability prediction model, BP neural network reliability prediction model and PSO-BP neural network reliability prediction model constructed were compared and verified through the actual data collected. The simulation results show that the average relative error of the reliability prediction model of the IPSO-BP neural network is reduced by about 17.5% and 10% compared to the average relative error of the reliability prediction model of the BP neural network and the PSO-BP neural network, respectively. This improves the accuracy of the reliability prediction of the traction inverter system of subway trains. At the same time, it has an important reference value for formulating a preventive maintenance cycle.