Very short-term surplus energy forecasting for peer-to-peer energy trading in microgrids

Tharumaraja, Sajarupan, Liyanage, Madura Prabhani Pitigala, Wijethunge, Akila, Ekanayake, Janaka ORCID: https://orcid.org/0000-0003-0362-3767 and Rupasinghe, Akbo 2025. Very short-term surplus energy forecasting for peer-to-peer energy trading in microgrids. Presented at: 5th International Conference on Advanced Research in Computing (ICARC), Belihuloya, Sri Lanka, 19-20 February 2025. https://ieeexplore.ieee.org/xpl/conhome/10962799/proceeding. IEEE, pp. 1-6. 10.1109/icarc64760.2025.10963019

Full text not available from this repository.

Abstract

Integrating renewable energy into microgrids facilitates peer-to-peer (P2P) energy trading between prosumers and consumers. However, variability in renewable power generation and dynamic consumer demand pose challenges to grid stability and trading mechanisms. This paper presents a forecasting method for available power from renewable-integrated households, focusing on Very short-term surplus power prediction for trading within a microgrid. The approach uses Internet of Things (IoT) monitoring, Long Short-Term Memory (LSTM) forecasting models, and the OpenDSS platform. The system includes an IoT-Photovoltaic (PV)-Grid Monitoring Unit to track power and indoor temperature, and an IoT-based Irradiance and Temperature Monitoring Unit for outdoor conditions. The monitored data is input into the LSTM model, and forecasted data is used in the OpenDSS DemandSide Management System (O-DSM) to predict PV generation and load behaviour in the microgrid at one-minute intervals over 15 minutes. The simulation provides insights into microgrid operations, energy flow, voltage profiles, and surplus power availability for trading, while the forecasting data supports trading decisions with peers. This research demonstrates the use of lightweight IoT devices and a cost-effective Raspberry Pi 5-based forecasting unit to enhance sustainable energy trading through improved forecasting and real-time data integration.

Item Type:	Conference or Workshop Item (UNSPECIFIED)
Date Type:	Published Online
Status:	Published
Schools:	Schools > Engineering
Publisher:	IEEE
ISBN:	9798331530990
Last Modified:	09 May 2025 11:45
URI:	https://orca.cardiff.ac.uk/id/eprint/178193

Actions (repository staff only)

Edit Item