摘要
针对地铁牵引逆变系统的可靠性预测问题,提出一种利用灰色理论优化BP神经网络的可靠性预测模型,充分发挥灰色理论需要数据少以及BP神经网络非线性拟合能力强的的特点,通过灰色理论构建出BP神经网络的拓扑结构并确定相关系数,采用BP神经网络对预测模型输出预测值的误差进行调整,并利用实际数据对改进前后的模型性能进行了对比。结果表明优化后的模型降低了数据处理带来的误差,预测精度明显提高,平均绝对百分比误差降低了约10.4%,对地铁牵引逆变系统的可靠性预测及制定维修维护策略具有重要的参考价值。
Aiming at the reliability prediction problem of the subway traction inverter system,a reliability prediction model based on grey theory was proposed to optimize the BP neural network,which fully exploited the characteristics of the gray theory and the strong nonlinear fitting ability of the BP neural network.The topological structure of BP neural network was constructed by grey theory and the correlation coefficient was determined.BP neural network was used to adjust the error of the predicted value of the prediction model output.The actual data were used to compare the performances of the model before and after the improvement.The results show that the optimized model reduces the error caused by data processing,the prediction accuracy is obviously improved,and the average absolute percentage error is reduced by about 10.4%.The reliability prediction and development of maintenance and repair strategies have important reference value.
作者
褚敏
李小波
王睿轶
王泉
CHU Min;LI Xiao-bo;WANG Rui-yi;WANG Quan(School of city rail transportation,Shanghai University of Engineering Sciences,Shanghai 201620,China;Shanghai Metro IT Co.,Ltd.,Shanghai 200233,China)
出处
《计算机仿真》
北大核心
2020年第7期168-171,共4页
Computer Simulation
关键词
地铁
牵引逆变系统
可靠性预测
灰色理论
神经网络
Subway
Traction inverter system
Reliability prediction
Grey theory
Neural network
