Enhancing Digital Autonomy in Peer-to-Peer Energy Trading: A Blockchain and Predictive Analytics Approach

Abstract

Distributed power generation from renewable energy sources is rapidly expanding and is serving as a clean energy alternative to fossil fuel-based energy sources, which pose significant environmental and health hazards. However, individuals still lack access to electricity and experience unreliable power supplies and services as demand increases. Energy distribution is still centralized, resulting in various challenges for users and contributing to disparities in energy distribution among individuals. To address these challenges, the predictive analytics approach and blockchain are employed together, while maintaining digital control and privacy for all participants in the decentralised energy trading system. The users can utilize their own historical energy consumption data to predict their own future energy requirements for the next period. Also, the prosumers (who both produce and consume energy) can predict their renewable energy generation. After analysing their consumption and generations users can use smart contracts to schedule and transfer energies over the blockchain network. This system guarantees transparency and security in energy trading between utility providers, prosumers, and consumers. The blockchain and smart contracts enables energy to be distributed fairly and automatically. This decentralization ensures digital autonomy within the energy trading system by enabling users to set their trading preferences, control their energy resources and data independently. This system with blockhain in energy trading builds trust among the participants. This proposed energy trading system is evaluated using a real world energy consumption and generation time series dataset of residential households. The proposed models for prediction are developed using Recurrent Neural Networks (RNN), such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). The results of both the models are compared using performance metrics such as mean absolute error (MAE) and root mean square error (RMSE). The results of experimental evaluation show that this approach can ensure reliable and equitable energy distribution, while ensuring digital autonomy and sovereignty in the proposed energy trading system.