Quantitative evaluation measures for DC-DC converters

Abstract

The use of power electronics has increased in everyday applications such as portable devices, biomedical, electric vehicles, etc. One of the major issues in such devices is an overshoot observed in the output voltage. The desired output in the converters comes with the over/undershoot problem which affects their performance. State-of-the-art evaluation measures such as overshoot percentage, settling time, and rise time only analyze a certain aspect and do not estimate the true performance of the system. In this work, we propose two new evaluation measures to analyze the performance of the converters overcoming the limitations of existing measures, i) Total over-under shoot (TS) and ii) First shoot second shoot (FSSS). We also propose one composite evaluation metric Aoun stability-1 to provide a single quantitative measure to analyze the performance of the system. A basic converter with a PID controller and Machine Learning based model is used to highlight the limitations of the existing metrics and strengths of the proposed metrics. Experimental results also highlight the advantages of the proposed measures. The proposed work is also expected to attract attention from the signal processing community facing similar issues in many applications.