Agarwal, Vaibhav ORCID: https://orcid.org/0000-0001-5362-2936 and Rani, Sujata (2026) Intelligent peer-to-peer energy trading for renewable smart grids using blockchain and optimized bidirectional recurrent neural networks. Computers and Electrical Engineering, 139 (Part B). p. 111422.

Full text not available from this repository.

Abstract

The rapid integration of renewable energy has accelerated the transition to decentralized smart grids. However, renewable intermittency, demand variability, and transaction security remain major challenges for efficient peer-to-peer energy trading. Therefore, intelligent forecasting and low-latency decentralized coordination are essential for reliable, efficient, and sustainable smart grid operation. This study proposes an intelligent, blockchain-supported framework titled XRP Ledger–Peer-to-Peer Energy Trading System using Bidirectional Gated Recurrent Neural Network and Parrot Optimizer (XRPL-P2P-ETS-BiGRNN-PO), which combines a Bidirectional Gated Recurrent Neural Network (BiGRNN) for accurate energy demand forecasting with the Parrot Optimizer (PO) for parameter tuning. Experimental results demonstrate that the proposed model outperforms existing methods, achieving 98.5% forecasting accuracy, an RMSE of 418, and an MAE of 283. It also records a transaction delay of 950 ms, a throughput of 1600 transactions per block, and carbon emissions of 962 tons, demonstrating a scalable, secure, and sustainable solution for smart grid energy trading.

Item Type:	Article
Status:	Published
DOI:	10.1016/j.compeleceng.2026.111422
Subjects:	T Technology > T Technology (General)
School/Department:	London Campus
URI:	https://ray.yorksj.ac.uk/id/eprint/15767

University Staff: Request a correction | RaY Editors: Update this record