Maximizing microgrid resilience: A two-stage AI-Enhanced system with an integrated backup system using a novel hybrid optimization algorithm

Abstract

This study focuses on the implementation of an efficient hybrid power system in two main phases. Firstly, it aims to optimize the performance of individual renewable energy sources (RESs), ensuring their maximum efficiency. Secondly, it employs artificial intelligence to enhance energy flow within the microgrid. The primary focus is on ensuring the reliability, sustainability, and cost-effectiveness of microgrid operations. In the first phase, the system focuses on optimizing power generation from PV and wind turbines (WTs). To maximize power extraction from the PV system, the study utilizes turbulent flow of water-based optimization (TFWO) and gorilla troops optimizer (GTO) algorithms. An advanced hybrid TFWO-ANFIS and hybrid GTO-ANFIS-based MPPT techniques are proposed, enhancing energy capture efficiency. These advanced MPPT techniques combine the adaptability of ANFIS with the optimization capabilities of metaheuristic algorithms, resulting in improved energy capture efficiency. Additionally, the system improves the performance of WTs by implementing the Zero d-axis current (ZDC) strategy. The second phase focuses on efficient energy storage and utilization of backup systems. It uses optimization algorithms to manage battery charging, hydrogen production, and the Vehicle-to-Grid (V2G) approach. The outcomes of the proposed study showcase optimized backup system operation, resulting in cost savings and increased utilization of RESs. The proposed system is comprehensively assessed across diverse operational scenarios. Four different cases are presented, each differing in the degree of involvement of backup sources. The results validate the effectiveness of the proposed strategy in managing power generation and load demand, ensuring optimal system performance. A comparative analysis of two algorithms is conducted, utilizing a set of 9 benchmark test functions. The results of this analysis highlight the superior speed and accuracy of TFWO in reaching the best solution. TFWO exhibited a faster convergence rate towards the optimal solution. The optimization algorithms succeed in reducing the operating cost significantly, notably reaching a 4.05 % reduction in the third scenario.