MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641009542 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Jan. 30, for 'an ai-driven ensemble framework for campus energy consumption forecasting and anomaly detection.'
Inventor(s) include Dr. K. Ragavan; Nithila Bhasker; Poshika Reddy; Rhythm Gupta; and Yami Sunil.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to an AI-driven ensemble framework for forecasting energy consumption and detecting anomalies across campus environments is disclosed. the framework comprises a data acquisition layer configured to ingest historical IoT-metered energy consumption datasets associated with multiple campus buildings, and a preprocessing and automated feature-engineering module that cleans, normalizes, and temporally structures the data while handling missing values. a multi-model prediction engine operates multiple predictive models in parallel, including at least one deep-learning model and one or more machine-learning models. a dynamic ensemble fusion engine adaptively weights model outputs based on validation performance metrics to generate robust forecasts. a residual-based anomaly detection engine identifies abnormal energy consumption patterns by analyzing deviations between predicted and observed values using adaptive statistical thresholds. an application and visualization layer provides forecasts, anomaly alerts, and actionable insights within a unified analytical pipeline. the framework operates as an automated decision-support system optimized for scalable, campus-level energy management without requiring real-time sensor connectivity to 2."
Disclaimer: Curated by HT Syndication.
