Optimal energy management for multi-energy microgrids using hybrid solutions to address renewable energy source uncertainty.

Research in industrial grid energy management is essential due to increasing energy demands, rising costs, and the integration of renewable sources. Efficient energy management can reduce operational costs, enhance grid stability, and optimize resource allocation. Addressing these challenges requires advanced techniques to balance supply, demand, and storage in dynamic industrial settings. In this study, a new hybrid algorithm is used for system modelling and low-cost, optimal management of Micro Grid (MG) networked systems. Optimizing micro-sources to reduce electricity production costs through hourly, day-ahead, and real-time scheduling was the process' primary goal.This research proposes a Quadratic Interpolation and New Local Search for Greylag Goose Optimisation (QI-NLS-G2O) and Gaussian Radius Zone Perceptron Net (GRZPNet) technique based energy management scheme for Multi-Energy Microgrids (MEMG) to help the Energy Management System (EMS) formulate optimal dispatching strategies under Renewable Energy Source (RES) uncertainty. To precisely represent the MEMG's operational state, the scheduling process incorporates an off-design performance model for energy conversion devices. Utilising MG inputs such as Wind Turbines (WT), Photovoltaic arrays (PV), and battery storage with associated cost functions, the GRZPNet learning phase based on QI-NLS-G2O is utilised to forecast load demand. The QI-NLS-G2O optimises the MG configuration according to the load demand. The MATLAB/Simulink working platform is used to implement the suggested hybrid technique, which is then contrasted with alternative approaches to solving problems.The proposed model significantly improves dispatching accuracy, reducing RES uncertainty impacts by approximately 15% and enhancing MEMG performance efficiency by up to 20% in simulations.