基于优化听觉模型的机床工况识别方法研究

准确识别机械制造设备工况,对判断设备的当前健康状态、设备平稳性,以及科学评价设备操作人员的工作效率具有重要意义.运用遗传算法优化后的ZCPA(zero crossings with peak amplitudes)听觉模型对设备的振动信号进行特征提取,通过与各种工况标准听觉谱计算相关性,以识别设备当前工况.优化后的ZCPA听觉模型计算简洁,能够模仿人耳听觉系统对输入信号进行提取,弥补传统听觉模型适应性差、识别率低的缺陷,同时使不同工况特征差异性增大,提高设备工况识别率.以某种普通车床为例,车床振动信号经过优化后的ZCPA模型处理后,工况识别率达到95%以上.

Method for the Condition Recognition of Machine Tool Based on an Optimized Auditory Model

The accurate identification of mechanical manufacturing equipment conditions is important for determining the current health status and stability of equipments and doing scientific evaluation of the operational efficiency of equipment operators.An improved zero crossings with peak amplitude (ZCPA) auditory model optimized by a genetic algorithm was used to extract the characteristics of the vibration signal of the device,and the current condition of the device was identified by the correlations analysis between various state standard auditory spectra.The optimized ZCPA auditory model can be used to imitate human auditory systems to extract the features of the input signal,to make up the defects of poor adaptability and low recognition rate of the traditional hearing model,and to increase the difference between different state features and improve the recognition rate of the equipment.Taking an ordinary lathe as an example,the condition recognition rate of the lathe vibration signal processed by the optimized ZCPA model reaches more than 95%.