【摘要】 归纳和总结了SOM神经网络多参数诊断法的实施步骤,阐述了轴承故障与振动信号之间的关系以及SOM神经网络的工作原理和实现过程,通过试验研究,提取了反映滚动轴承故障类型的振动信号的特征参数,以构建训练神经网络的特征向量,利用MATLAB 7.0人工神经网络工具箱(ANN)模拟和仿真SOM神经网络,然后用训练后的SOM神经网络对故障模式进行识别。更多还原

【Abstract】 It summarizes the steps of the multi-parameter diagnosis method based on Self-Organizing Feature Map neural network and illuminates the connection between fault and vibration signal of rolling bearings,and operating principle and implementation procedure of SOM neural network.Via experimental investigation,the characteristic parameters which are able to image the fault type of the rolling bearing are extracted to construct the characteristic vectors of training a neural network.The SOM neural network is simulated using ANN(Artificial Neural Network)toolbox of MATLAB7.0 and then the fault mode is identified with the trained SOM neural network.更多还原