Health Monitoring of an Isolated Flyback Converter Using Output Parameters

As DC-DC converters increasingly employed in various applications, the failure of such critical units emerges as a challenging problem and is becoming markedly prominent. In order to estimate the real time health status of such systems, this paper presents an output parameter based health monitoring method for a flyback DC-DC converter. Vulnerable components are identified and degradation models thereof are formulated. Subsequently, Monte Carlo simulation is performed considering component tolerance by means of LTspice. Health indicating parameters are selected through simulation result analysis. Taking the health parameters as inputs, a RBF-based health monitoring model is established. The proposed model performs well in health monitoring of the circuit with a high accuracy in health status estimation. The introduced approach can be further applied to other systems for health management.