Health-assessment methodology research for SMPS based on simulation

Abstract

& CO CLUSIO SAs the idea of "all-electric" has been proposed and gradually realized in recent years, the power consumption of the system increases substantially, this demands the power supply with better properties and higher reliability. Currently, the prognostics and health management research of the power electronic system, including switch mode power supply (SMPS), are based primarily on field data with specified usage stress. The usage stress (including input voltage, temperature, load, etc.) is specific, therefore, it is difficult to adapt to the measurement in other stress conditions. The tolerance of the system is rarely considered. In addition, the basis of the health-assessment is the database containing degradation features of multi-component degradation. However, the degradation degree of components can rarely be controlled by experiment. By injecting degradation parameters of components into the high fidelity SMPS model, a healthassessment methodology for SMPS based on simulation is proposed. Firstly, SMPS with multi-component degradation in various tolerance and stress conditions is simulated by using the design of experiment (DOE) method; Secondly, the testable signals in the system such as output signals and gatesource voltage of MOSFET are pre-treated to reduce dimensionality and to eliminate redundant information, and the database containing degradation characteristics is created. Then, based on the database, the degradation status of SMPS is identified by Mahalanobis Distance (MD) calculation. The methodology proposed in this paper provides a general platform for prognostics and health-management research of SMPS. A flyback SMPS is studied to prove the effectiveness and accuracy of proposed methodology.

RUL Prediction of Switched Mode Power Supply Using a Kalman Filter Assisted Deep Neural Network

“…Zhang et al proposed a four-step PHM approach of SMPS, which are precursor parameter identification, baseline establishment, baseline verification, and testing [27]. Xuerong Ye et al used a simulation-based health assessment methodology for SMPS [28]. Min Zhao et al made a proposal for failure prediction methodology for SMPS power MOSFET using an equivalent circuit model [29].…”