用于机械诊断的最佳特征参数自动生成法

A Method of Automatic Generation of Optimum Symptom Parameters for Diagnosis of Plant Machinery

将计算机用于机械工况的自动诊断时 ,必须从采集的信号中提取表征机械工况的特征参数 (简称SP) ,但目前尚无一种令人满意的提取最佳 SP的方法。为了克服这个困难并能保证对机械工况进行高精度的诊断 ,本文基于遗传基因算法 (简称 GA)提出了一种称为“特征参数自动生成”的新方法 ,能快速找到最佳 SP。

When using computer for automatic condition diagnosis of plant machinery, the symptom parameters (SP), by which the failure types of plant machinery can be sensitively distinguished, are indispensable. However, there is not an acceptable method for extracting optimum SP from signals measured in normal and abnormal states. In many cases, the practice of extracting SP is not only troublesome and labor intensive, but also can not ensure that the optimum SP will be extracted. Therefore, by using Genetic Algorithms (GA), a new method called “Automatic generation of symptom parameters” was proposed in order to generate the optimum SP automatically and insure highly accurate diagnosis. Applying the method to diagnosis of plant machinery, the optimum SP could be quickly generated. The diagnosis of a gear equipment as an example verified that this method is very effective.