MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641109627 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And Technology on September 12, 2026, for A Machine Learning-Based System For Predicting Smart Mix Proportions And Compressive Strength For One-Part Geopolymer Composites.
Inventors include Rayana Harika; Dr. Kolli Ramujee; and D. Kartheek.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: A computational system is provided for the prediction of compressive strength and generating suitable mix proportions for one-part geopolymer composite. The system processes material composition, activator content, curing conditions, aggregate proportions, and derived mix parameters obtained from geopolymer paste, mortar, and concrete data. Four machine learning models are used in the hybrid stacking framework, along with a Ridge Regression meta-learner, to generate the predicted compressive strength, and Gaussian Process Regression to derive the related prediction uncertainty. For mix generation, the compositions of candidates are compared to the required target strength and specified engineering constraints: such as material dosage limits, water/binder ratio, activator/binder ratio, and curing conditions. The selected mix proportions and prediction results are displayed via a web based graphical user interface titled “Smart Mix Generator”. Laboratory testing of selected one- part geopolymer mixes is used to compare measured compressive strengths with the corresponding model predictions.
Disclaimer: Curated by HT Syndication.
