A secure and energy-efficient IoT-blockchain framework for decentralized renewable energy trading

The concept of decentralized energy trading is transforming the multiple ways of trading renewable energy, and the conventional method that requires aggregators is hindering speed and reliability. Therefore, we have suggested a decentralized IoT-blockchain architecture with Convolutional Neural Networks (CNN)-based fraud detection and K-Means cluster to match the prosumers and consumer. Our framework succeeds in the transaction in 93.9 percent of cases compared to traditional aggregator-based trading platforms, which are characterized by a centralized system and delays in transactions, and achieve a higher fraud detection rate of 98.5. Also, it also improves energy distribution efficiency by 24.3% and network resilience by 17.6% and hence peer-to-peer markets can be made viable and secured. CNN model is used to identify anomalies (in real-time) as the clustering (best trade paths) is used to find the best trade paths based on demand profiles. To ensure the responsiveness, scalability, and security of the system, the simulations of trading and blockchain implementation scenarios were carried out in the MATLAB Simulink and Hyperledger Fabric. The current work has provided a more favorable platform to the decentralized paradigm of energy exchange by providing an intelligent, a faster and a safer model as compared to the traditional systems that were centralized around aggregators. • A propose decentralized IoT-blockchain framework with Convolutional Neural Networks (CNN)-based fraud detection and K-Means clusters for matching the prosumers and consumer. • It also increases energy distribution efficiency by 24.3% and network resilience by 17.6%, thus making peer-to-peer markets viable and secured • The present work has set a better stage for the decentralized energy exchange paradigm by introducing an intelligent, faster, and safer model