一种基于Rough集和RBF网络的模拟电路故障诊断方法

A method based on rough set and RBF neural network for analog circuit faults diagnosis

提出了一种基于Rough集和RBF神经网络结合的故障诊断方法。该方法针对模拟电路的故障特征,进行粗集约简预处理研究,然后由约简后的属性构造RBF网络。仿真结果表明:在相同的精度要求下,该算法的训练时间远小于普通的进化神经网络,提高了泛化能力,对模拟电路的故障诊断有一定的实际意义。

A method based on the combination of rough set and Radial Basis Function neural Network is presented for analog circuit faults diagnosis. Directing to the fault characteristic of analog circuit, this paper has proposed rough set reduction pretreatment, the reduced decision attributes are used to construct RBF neural network. The simulation result indicated that, under the same precision request, this algorithm training time far is smaller than the ordinary evolution neural network and improve the ability of generality.