Support Vector Neural Network and Principal Component Analysis for Fault Diagnosis of Analog Circuits

Abstract

Fault diagnosis of the analog circuits is the trending research area as the analog circuits holds a lot of applications in military, automatic control, household appliances, communication, and so on. Even though the researchers presented various methods for fault diagnosis, still there is a lack of reliable techniques for analog fault detection and diagnosis. Keeping this mind, this paper presents the Support Vector Neural Network (SVNN) for identifying the faulty and the fault-free analog circuit. At first, the pre-processing is carried out using the Principle component analysis (PCA) that serves as the best way for solving the dimensional complexities. Then, the weights of SVNN are optimally tuned using the Genetic Algorithm (GA) that enables the optimal classification of the analog circuits. The GA-based SVNN is an optimization approach for classifying the analog circuits that enable the comprehensive diagnosis of the faults in the analog circuits. The experimentation is performed using the triangular wave generator and the simulation results highlight that SVNN classifier attained a classification percentage of 99.54% and low False Alarm Rate of 0.68%.