Optimizing Two-stage Energy Management in renewable-based Multi-Microgrid using a Modified Student Psychology-Based Optimization with Demand Response and Hydrogen Storage

Hydrogen energy storage systems (HESS) play an essential role in Microgrid (MG) systems to address the inherent generation characteristics of renewable energy sources. Also, the integration of Demand Response (DR) into the Energy Management System (EMS) of a renewable -based Multi-Microgrid (MMG) can lead to substantial technical and economic benefits. This paper proposes a modified optimization algorithm for optimizing MMG Energy Management (EM). The proposed algorithm is a modified version of the Student Psychology -Based Optimization (SPBO) technique called Modified Student Psychology -Based Optimization (MSPBO). This modification aims to improve issues such as slow convergence, low solution accuracy, lack of diversity, and getting stuck in local optima. The proposed MSPBO method incorporates a local escape operator and a collaborative student class to achieve a better balance between exploiting known solutions and exploring new possibilities. The MSPBO algorithm is applied to address the EM challenge within a MMG context. Considering the integration of renewable sources as Wind turbines and solar photovoltaic and the HESS, the EM problem is formulated as a two -stage multi -objective optimization: minimizing the operating cost of conventional generators and power transactions cost in addition to the cost of HESS, while maximizing operator benefits and peak load reduction. This multi -objective problem is tackled using a hybrid epsilon-lexicography - weighted-sum approach that avoids the need for normalization. The performance of the proposed MSPBO is evaluated using CEC 2017 benchmark test functions, utilizing various statistical measures such as best, average, worst, rank, and standard deviation (SD) of fitness values, along with Wilcoxon ' s rank -sum test. The MSPBO technique is compared with other optimization algorithms for these test functions, highlighting its efficiency and adeptness in achieving a harmonious trade-off between exploiting known solutions and exploring new ones. Furthermore, the MSPBO method is applied to solve two case studies. In Case 1, which involves a single stage with conventional demand response optimization, the results achieved using MSPBO are benchmarked against other optimization techniques, revealing its superior efficacy in addressing the EM challenge. In Case 2, a more complex two -stage multi -objective problem is tackled using MSPBO, and the assessment of the integration of HESS in the MMG is evaluated. In the second stage, there is a notable enhancement in peak load reduction percentage (PRP), from 13.9% to 16.13% without the HESS and from 12.68% to 16.46% with the integration of HESS.