The economics, reliability, and carbon efficiency of hybrid microgrid systems (HMSs) are often in conflict; hence, a reasonable design for the sizing of the initial microgrid is important. In this article, we propose an improved two-archive many-objective evolutionary algorithm (TA-MaEA) based on fuzzy decision to solve the sizing optimization problem for HMSs. For the HMS simulated in this article, costs, loss of power supply probability, pollutant emissions, and power balance are considered as objective functions. For the proposed algorithm, we employ two archives with different diversity selection strategies to balance convergence and diversity in the high-dimensional objective space. In addition, a fuzzy decision making method is proposed to further help decision makers obtain a solution from the Pareto front that optimally balances the objectives. The effectiveness of the proposed algorithm in solving the HMS sizing optimization problem is investigated for the case of Yanbu, Saudi Arabia. The experimental results show that, compared with the two-archive evolutionary algorithm for constrained many-objective optimization (C-TAEA), the clustering-based adaptive many-objective evolutionary algorithm (CA-MOEA), and the improved decomposition-based evolutionary algorithm (I-DBEA), the proposed algorithm can reduce the system costs by 7%, 13%, and 21%, respectively.

中文翻译：

---

基于模糊决策的混合微电网多目标选型优化

混合微电网系统 (HMS) 的经济性、可靠性和碳效率经常相互矛盾；因此，合理设计初始微电网的规模很重要。在本文中，我们提出了一种基于模糊决策的改进的双归档多目标进化算法（TA-MaEA）来解决 HMS 的尺寸优化问题。对于本文模拟的HMS，以成本、停电概率、污染物排放和功率平衡为目标函数。对于所提出的算法，我们采用了两个具有不同多样性选择策略的档案来平衡高维目标空间中的收敛性和多样性。此外，提出了一种模糊决策方法，以进一步帮助决策者从帕累托前沿获得最优平衡目标的解决方案。以沙特阿拉伯延布为例，研究了该算法在解决 HMS 尺寸优化问题中的有效性。实验结果表明，与约束多目标优化的二归档进化算法（C-TAEA）、基于聚类的自适应多目标进化算法（CA-MOEA）和改进的基于分解的进化算法相比(I-DBEA)，所提出的算法可以分别降低系统成本 7%、13% 和 21%。