自适应免疫遗传BP神经网络的高压发泡机故障诊断  

Fault Diagnosis of High Pressure Foaming Machine Based on Adaptive Immune Genetic BP Neural Network

作　　者：梁宝峰 张永林[1] LIANG Baofeng;ZHANG Yonglin(College of Electronics and Information,Jiangsu University of Science and Technology,Zhenjiang 212003)

机构地区：[1]江苏科技大学电子信息学院,镇江212003

出　　处：《计算机与数字工程》2020年第9期2289-2294,共6页Computer & Digital Engineering

摘　　要：为提高聚氨酯高压发泡机故障诊断系统的准确率,将免疫遗传算法改进的BP神经网络诊断方法应用于聚氨酯高压发泡机系统。选取BP神经网络的误差函数倒数作为遗传算法的适应度函数,在遗传算法中加入免疫算子,免疫算法疫苗提取、接种环节和遗传算法的交叉、变异环节采用自适应控制策略,最后把算法最优解赋给BP神经网络的权值和阈值,用实测数据分别对改进的BP神经网络诊断系统和原有的BP神经网络诊断系统进行Matlab仿真对比。仿真结果表明,优化后的故障诊断系统有更高的精度和诊断效率,诊断准确率达到96.7%。In order to improve the accuracy of fault diagnosis system for polyurethane high-pressure foaming machine,the improved BP neural network diagnosis method based on immune genetic algorithm is applied to polyurethane high-pressure foaming system.The reciprocal error function of BP neural network is selected as the fitness function of genetic algorithm.The immune operator is added to the genetic algorithm.The adaptive control strategy is adopted in the immune algorithm vaccine extraction,vaccination and genetic algorithm crossover and mutation.Finally,the optimal solution of the algorithm is assigned to the weights and thresholds of BP neural network.The improved BP neural network diagnosis system and the original BP neural network diagnosis system are simulated with the measured data by Matlab.The simulation results show that the optimized fault diagnosis system has higher accuracy and diagnostic efficiency,and the diagnostic accuracy rate reaches 96.7%.

关 键 词：高压发泡机 BP神经网络 免疫遗传算法 自适应控制策略 故障诊断 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]