To improve the diagnosis accuracy of tolerance fault for analog circuits, a Wavelet Neural Network (WNN) diagnosis method based on a modified UKF algorithm is proposed on the basis of feature extraction. An adaptive factor based on variance inflation principle is firstly introduced to improve the performance of UKF algorithm. Then, the modified UKF algorithm is used to perform optimal estimation for WNN parameters, establishing the tolerance fault diagnosis model based on feature extraction. The simulation results on Sallen-Key band-pass filter show that, the proposed method has a good convergence rate and diagnosis correct rate, which validates the feasibility and effectiveness of this method.
Tolerance Fault with Variance Inflation Factor for Analog Circuit
2019-10-01
64761 byte
Conference paper
Electronic Resource
English
Space shuttle fault tolerance: analog and digital teamwork
Tema Archive | 2009
|Analog Circuit Fault Diagnosis with Multi-Objective Particle Swarm Optimization
British Library Online Contents | 2012
|