基于概率模型的遗传规划在故障诊断中的应用  

Fault detection using genetic programming with probabilistic model

作　　者：陈东用[1] 梁艳超[2] 周永权[3] 李陶深[1] 

机构地区：[1]广西大学计算机与电子信息学院,广西南宁530004 [2]哈尔滨工业大学机电工程学院,黑龙江哈尔滨150001 [3]广西民族学院计算机与信息科学学院,广西南宁530006

出　　处：《计算机工程与设计》2006年第22期4210-4212,共3页Computer Engineering and Design

基　　金：国家自然科学基金项目(60461001)

摘　　要：采用在遗传规划中使用概率模型的新方法来解决一系列故障诊断问题。故障诊断可被看为是一个多级分类问题。遗传规划在解决复杂问题上有很大的优势,而这种优势在故障诊断中仍然显著。而且,使用概率模型作为适应函数能提高诊断的精确性,最后用这种方法解决机电设备的故障诊断。结果显示,使用基于概率模型的遗传规划解决机电设备的故障诊断比人工神经网络优越。A new method is presented to solve a series of fault detection problems using probabilistic model （PM） in genetic programming （GP）. Fault detection is seen as a problem of multi-class classification. GP methods used to solve problems have a great advantage in their power to represent solutions to complex problems, and this advantage remains in the domain offault detection. Moreover, diagnosis accuracy is improved by using PM. In the end, the fault detection of electro-mechanical device is solved. The results show that the method using GP with PM to solve fault detection of electro-mechanical device is superior to the artificial neural network （ANN）.

关 键 词：概率模型 遗传规划 故障诊断 机电设备 人工神经网络 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]