MUMBAI, India, Sept. 7 -- Intellectual Property India has published a patent application (202641064121 A) filed by Srm Institute Of Science And Technology Tiruchirappalli on May 21, 2026, for Machine Learning-Based Efficiency Prediction System For Dc–dc Power Converters.
Inventors include Akaash G; Dharsan B; and Dr. Shajith Ali.
The application for the patent was published on September 04, 2026, under issue no. 36/2026.
Abstract: The present invention provides a machine learning–based system for predicting the efficiency of DC–DC power converters under varying operating conditions. The invention integrates power electronics modeling with artificial intelligence techniques to develop a data-driven framework for efficiency estimation. Instead of relying on complex analytical loss calculations and repeated experimental testing, the proposed system utilizes simulation-generated or measured datasets to train a machine learning model that learns the nonlinear relationship between key operating parameters and converter efficiency. The system accepts fundamental operating inputs such as input voltage, duty cycle, load resistance, and switching frequency, and processes them through a trained regression-based model to predict efficiency instantly. By eliminating the need for explicit mathematical modeling of switching and conduction losses during prediction, the invention significantly reduces computational effort and simplifies converter performance evaluation. The proposed approach enables rapid efficiency estimation during the design phase, supports optimization studies, and enhances ease of use for engineers and researchers.
Disclaimer: Curated by HT Syndication.
