Research On Abnormal Power Condition Monitoring Of Batch Production In Numerical Control Machine

In order to guarantee the safety of production devices and the quality of production in the modern manufacturing system, there's critical need to monitor the production process. One of the most important factors in machine tool production is ensuring the machine tools work well. In the variable-frequency machine tool, when process the same work piece, the power information is related stable. But dilapidation and abrasion of machine tools, or abrasion of spindle can effect the power information that the real-time power information will be different from the power information of normal production. Among the available monitor methods, the power monitoring takes advantage over the others through its simplicity and practicality. But there're also some disadvantages in power monitoring. The paper improves the delicacy and universality through wavelet transform and cutting correlation detection.At first, the reason why variable-frequency machine tool has stable power information is introduced. This paper introduces and compares some common machine tools monitoring methods. Some improvement is mentioned in this paper, in order to overcome lack of delicacy and difficulty of ascertaining threshold value of traditional power monitoring method.Then, the stability of wavelet transform is introduced because that the traditional power monitoring method lacks delicacy in the power signal. This part of the paper focus on the advantage, disadvantage and universality of the Mallat algorithm. We improve the ordinary wavelet transform algorithm. The improved wavelet transform algorithm meets the need of real-time processing and reduces the time complexity. Correlation detection is put forward to overcome the weakness that the traditional power monitoring is difficult to ascertain the threshold value. Because of the need of real-time processing, necessary improvement to the correlation detection is needed. The search of peak value is a key element in correlation detection..At last, the paper shows the material way to integrate the wavelet transform and correlation detection to the power monitoring. The result of the improved real-time power monitoring of machine tools is presented production of practical workpiece.