基于退火遗传混合算法的模拟电路诊断激励优化

本文提出基于Volterra核和模拟退火遗传混合算法的模拟电路故障诊断激励优化方法.在以Volterra核为特征向量的模拟电路故障诊断中,以相同激励信号下电路各故障状态的特征向量的集总欧氏距离作为适应度函数,对用于激励的多频正弦信号的参数进行优化,首先利用模拟退火算法形成精英团队,然后利用遗传算法寻找最佳激励信号的参数,从而提高故障诊断的效率.文中给出了退火遗传混合算法的优化方案和流程,并通过实例加以验证.

Optimizing the Stimulus of Analog Circuits Diagnosis Based on Simulated Annealing and Genetic Algorithm

The optimization and fault diagnosis method of analog circuit based on Volterra kernel and simulated annealing genetic hybrid algorithm is presented. In a analog circuit fault diagnosis which use the Volterra kernel as feature vector,make the characteristic vector lumped Euclidean distance of serials of fault state under the same excitation signals as the fitness function,and optimize the parameters used to stimulate the multi-frequency sinusoidal signal. Firstly,use simulated annealing to form the elite te...