SECURE ENERGY DEMAND PREDICTION FOR ELECTRIC VEHICLES USING FEDERATED LEARNING ON BLOCKCHAIN

Abstract

This investigation employs blockchain technology and federated learning to develop a novel approach to accurately forecast the energy requirements of electric vehicles (EVs). It has become increasingly challenging to accurately forecast the amount of energy required while simultaneously safeguarding data privacy and security as more individuals adopt electric vehicles. Federated learning safeguards user privacy by enabling energy providers and electric car manufacturers to collaborate in the development of predictive models without transmitting raw data to one another. The decentralized ledger and smart contracts of blockchain technology simultaneously safeguard data security, transparency, and trust. The comprehensive approach effectively anticipates fluctuations in energy consumption while simultaneously addressing significant challenges related to scalability, security, and privacy. The results of the experiments indicate that this architecture is effective in safeguarding private data from intruders and enabling smart grids to manage their energy in a reliable and long-lasting manner. This research offers a robust approach to enhance the safety and utility of EV energy demand forecasting.