A Method of Fault Diagnosis Used For Circuit Blocks Based on Adaptive Neural Network

摘要

The general Back-propagation algorithm has some disadvantages, such as the speed of learning convergence is too slow and local extreme values are present in the process of search sometimes. In order to solve the problems described above, this paper applies adaptive learning coefficient and momentum term to improve the general Back-propagation algorithm. Taking the example of the phased array radar circuit blocks' fault diagnosis, the basic fault features are transformed by means of using the fuzzy sets of fuzzy mathematics, and take them as the inputs of adaptive neural network. And then, the fault codes are used as the outputs of adaptive neural network. The experiment simulation results show that adaptive neural network has many advantages, such as the process of learning is very capable, the time of search is short, and the operational precision is high, etc. So we get a conclusion that this method can diagnose and identify the fault types of phased array radar circuit blocks effectively.