MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611116811 A) filed by Lovdeep Grover; Prof Narayan Malakar; Dr Gautham Krishna; Dr. Hema Kale; Manoj Bechooprasad Maurya; Dr. Puneet Kumar; Dr. Kirti S. Vaidya; Dr Algubelly Yashwanth Reddy; Rajput Jayasree; Sneha Baddayyagari; S. Jeyapriyanga; and Polu Veeraraghava Reddy on September 30, 2026, for Machine Learning-Based Adaptive Energy Management System For Solar-Powered Smart Cities.
Inventors include Lovdeep Grover; Prof Narayan Malakar; Dr Gautham Krishna; Dr. Hema Kale; Manoj Bechooprasad Maurya; Dr. Puneet Kumar; Dr. Kirti S. Vaidya; Dr Algubelly Yashwanth Reddy; Rajput Jayasree; Sneha Baddayyagari; S. Jeyapriyanga; and Polu Veeraraghava Reddy.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention discloses the development of an adaptive energy management system (100) and methodology (200) using machine learning algorithms for solar- powered smart cities is provided. The proposed system (100) makes use of a plurality of IoT sensors (103) to collect real-time weather and load data that is processed by an edge computing gateway (104) and then communicated to the ML engine (105) on cloud computing infrastructure. The ML engine (105) uses a Long Short-Term Memory neural network (108) in the predictive forecasting module (106) to perform solar generation and load prediction. Additionally, a Deep Reinforcement Learning algorithm (111) residing in the optimization module (107) optimizes the real-time energy dispatching policy (113). The Deep Reinforcement Learning algorithm (111) operates in the digital twin environment (112) to maximize solar self-consumption, minimize grid dependency, and keep batteries in good condition. A programmable logic controller (114) implements the optimized energy dispatching policy (113) by controlling bi-directional inverters (116) and smart grid tie-breakers (117). FIG.1
Disclaimer: Curated by HT Syndication.
