基于WPD_EMD和SVM刀具磨损故障诊断模型

Tool Wear Fault Diagnosis Model Based on WPD_EMD and SVM

通过采集2种磨损程度不同的同类型刀具加工工件时机床主轴的振动信号,提出WPD_EMD和SVM故障诊断模型判断刀具磨损程度。首先利用小波包工具去除高频噪声信号,其次利用EMD分解得到若干个固有模态函数和一个残差,计算各个固有模态函数和EMD分解前信号的相关系数,合并相关系数大的固有模态函数得到新信号。计算新信号的绝对均值作为时域特征参数。选取若干组试验数据作为支持向量机训练集,建立判断刀具磨损程度大小的故障诊断模型。试验表明该故障模型预测刀具磨损程度准确率100%,为判断刀具实时加工工件的磨损程度提供新的途径。

The vibration signal of the spindle of the machine tool is collected when different wear degree of two same type cutting tools is cutting the same workpiece. The WPD_EMD and SVM fault diagnosis model is used to judge the tool wear degree. Firstly, the wavelet packet decomposition is used to remove the high frequency noise signal. Secondly, a number of intrinsic mode functions and a residual error are obtained by using EMD decomposition. The correlation coefficient of each intrinsic mode function and signal which is not used EMD decomposition is calculated. A new signal is got by adding together the intrinsic mode functions with large correlation coefficient. The absolute mean value of the new signal is used as the characteristic parameter in the time domain. The support vector machine training set is selected from some group of test data. The tool wear fault diagnosis model is established. The accuracy rate of tool wear fault diagnosis model is 100% through experiments. A new way is provided to judge whether the workpiece is good when wearing the workpiece.