Real-Data based Economic Emission Load Dispatch with Renewable Energy and Electric Vehicle Integration using Artificial Ecosystem-based Optimization

Abstract

This paper presents a data-driven approach to the Economic Emission Load Dispatch (EELD) problem that integrates Renewable Energy Sources (RES) and Plug-in Electric Vehicles (PEVs). The Artificial Ecosystem-Based Optimization (AEO) algorithm is used to address the stochastic nature of the power system of the future by including real wind and solar data from the ‘Renewable.ninja’ platform for Gujarat, India. This data-driven framework not only captures essential uncertainties in RES generation but also the arrival, departure, and waiting times for PEVs. The method uses the AEO algorithm to simulate ecosystem-like interactions that help the system achieve a good balance between exploration and exploitation, thereby minimizing both total generation costs and environmental emissions. The study has been conducted on 10-unit and 20-unit thermal generation systems, including practical Vehicle-to-Grid (V2G) and Grid-to-Vehicle (G2V) operations. The main results reveal that the AEO algorithm is instrumental in improving system performance by offering a good balance of trade-off between economic and environmental goals. Also, a performance comparison of the AEO algorithm with the latest optimization methods shows that AEO is more effective and stable under complex dispatch scenarios. The paper argues that combining real-world data with ecosystem-based optimization not only provides a scalable, sustainable solution but also offers a way out of modern grid management. This study stands out for combining site-specific meteorological data with high-fidelity PEV behavioural modelling to provide a practical, field-ready strategy for utility operators to manage the volatility of green energy and electric mobility transitions.