混沌神经网络在球磨机故障诊断中的应用

Application of chaos neural network in fault diagnosis of grinding machine

针对神经网络极易陷入局部极小的问题,采用引入动量项和混沌映射的改进BP算法,讨论引入动量项和混沌映射的神经网络综合模型的建模思路及其算法实现;介绍球磨机常见的故障类型,建立球磨机故障诊断的混沌神经网络模型,进行仿真试验,结果表明:该模型具有较高的预测精度,可以有效地运用于球磨机诊断中的故障预测。

The improved BP algorithm with added momentum item and chaotic mapping is proposed to overcome the problem in make-up local minima of neural network . The idea of building model based on improved BP algorithm with added momentum item and chaotic mapping is discussed. The types of faults for grinding machine are described. The choas neural network model in fault diagnosis of grinding machine is built and the simulation test is made. The result shows that the model has a high accuracy and can effectively forecast the fault in diagnosis of grinding machine.