MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641096431 A) filed by Sathishkumar Nagarajan; Dr. Manam Ravindra; Dr. Harsha Rathore; Elango Ponnusamy; Krishnamoorthy N; Karthik Annamalai; Dr. N. P. Gopinath; Dr. P. Senthilkumar; and Dr. John De Britto C on August 10, 2026, for Reinforcement Learning Based Ev-To-Grid Energy Scheduling Mechanism.
Inventors include Sathishkumar Nagarajan; Dr. Manam Ravindra; Dr. Harsha Rathore; Elango Ponnusamy; Krishnamoorthy N; Karthik Annamalai; Dr. N. P. Gopinath; Dr. P. Senthilkumar; and Dr. John De Britto C.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: Abstract The present invention recounts to a Reinforcement Learning (RL)-based Electric Vehicle-to- Grid (EV-to-Grid or V2G) energy scheduling instrument intended to augment bidirectional energy amid electric vehicles, renewable energy sources, battery energy storage systems, and the-utility-grid; The planned system documents electric vehicles to work as distributed energy storage units talented of perceptively indicting since or settling to the grid based on real-time working situations. Dissimilar conventional rule- based or optimization-based scheduling means, the RL agenda uninterruptedly absorbs from active conservational situations, counting electricity pricing, renewable energy convenience, battery state-of-charge, user mobility necessities, and grid claim, to govern ideal charging and discharging policies. The RL manager services a reward-based learning course to exploit renewable energy utilization, diminish electricity costs, diminish peak load request, augment battery lifespan, and progress overall grid stability. The planned planning partakes solar photovoltaic systems, wind energy, bidirectional power converters, battery management systems, smart meters, foT-enabled statement, and cloud-based energy management for uninterrupted monitoring and smart control. Additionally, renewable energy forecasting and load prediction are combined to advance scheduling accuracy and system reliability. The conception chains demand reply, frequency regulation, top shaving, gorge filling, and backup stoppage power while guaranteeing battery health guard and user mobility chains. Through adaptive policy tells and nonstop contact with the operating situation, the planned mechanism eliminates the requirement for predefined precise models and physical expansion. The Reinforcement Learning-based EV-to-Grid Energy Scheduling Mechanism offers an intelligent, walkable, and sustainable reply for imminent smart grids and electric mobility biology by calming energy efficiency, plunging carbon emissions, pretty renewable energy integration, and compressing power system suppleness
Disclaimer: Curated by HT Syndication.
