粗糙集理论在智能故障诊断中的应用

Applications of Rough Set Theory in Intelligent Fault Diagnosis

神经网络和规则推理是智能故障诊断的两种重要方法。给出了粗糙集理论在这两种系统中的应用 :粗糙集—神经网络 (RNN)系统。粗糙集作为神经网络系统的预处理 ,仿真结果表明 RNN系统提高了诊断准确率和诊断速度 ;粗糙集用于故障诊断专家系统的规则获取 ,可得出确定性规则和可能性规则。结果表明粗糙集方法能处理由于类重叠引起的样本信息不精确、不一致情况下的规则获取 ,消除故障诊断中的误报和漏报现象对诊断性能的影响。

Neural network and rule reasoning are two important methods of intelligent fault diagnosis. In this paper we give two applications of Rough Set in the fault diagnosis:(1) Rough Set-Neural Network(RNN)system, i.e. Rough Set acts as preprocessing of the Neural Network system,the simulation results indicate that the RNN system has increased correct rate and velocity of diagnosis. (2) Rough Set is used to rule acquisition of fault diagnosis expert system, and can elicit certain rules and possibility rules. The results of this paper indicate that the Rough Set method can solve the problem of rule acquisition under the condition of imprecise and inconsistent information on account of kind overlapping, and can eliminate the effects of misinformation and failing to report on the quality of diagnosis in fault diagnosis.