The invention discloses a method for optimizing analog circuit measuring points based on a
sawtooth wave and a
genetic algorithm. The method for optimizing the analog circuit measuring points based onthe
sawtooth wave and the
genetic algorithm comprises the following steps that fault
simulation is carried out on each
fuzzy group representative element of each element of the analog circuit for a plurality of times, and the
sawtooth wave is used as an input of the analog circuit during analog to obtain fault voltages corresponding to the measuring points under different frequencies, so that analog circuit data are obtained; individuals in the
genetic algorithm are initialized, the individuals are processed, then the
feature vector of the corresponding measuring points from the analog circuit data is screened according to a measuring point selection scheme corresponding to the individuals, cross-
verification is carried out on a preset classifier, the obtained classification accuracy rateis used as a fitness value of the individuals, then a next generation
population is generated to continue
processing until an iteration end condition is reached, and the individual with the greatestfitness in the current
population is used as the optimal individual, and the measuring point selection scheme corresponding to the individual is a measuring point optimizing scheme. The method for optimizing the analog circuit measuring points based on the sawtooth wave and the genetic
algorithm can effectively realize the measuring point optimizing and improve the accuracy of fault diagnosis whenanalog circuit fault diagnosis is carried out based on multiple frequencies.