Abstract: To address the challenges of operational safety in hydrogen production equipment within electric-thermal-hydrogen multi-carrier energy systems (MES), as well as the difficulty in balancing economic efficiency and low-carbon operation in dispatching, this study proposes an optimized scheduling strategy that combines a nonlinear dynamic model of the electrolyzer with the Non-dominated Sorting Genetic Algorithm-II (NSGA-II) for multi-objective optimization of the MES. First, to ensure the hydrogen production process operates within a safe temperature range, a refined dynamic model of an alkaline electrolyzer is developed, accounting for the nonlinear variation in electrolysis efficiency due to stack temperature fluctuations. Second, to leverage the strong coupling relationships among various forms of energy and the dispatchability of loads within the MES, an integrated demand response mechanism and carbon trading mechanism are introduced to enhance the system’s economic efficiency and low-carbon operational capability. Finally, by integrating the complex nonlinear model of the electrolyzer, a multi-objective optimization algorithm based on NSGA-II is proposed, with the optimal solution evaluated comprehensively through the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). Simulation experiments across different scenarios compare the system's daily operating costs and carbon emissions. The results of the study indicate that the multi-objective optimization method combined with NSGA-II can significantly reduce the daily total operating cost and carbon emissions of the system. Compared to optimizing only the economic cost, the daily total cost and carbon emissions are reduced by 33.4% and 57.5%, demonstrating the proposed strategy's effectiveness in enhancing economic efficiency, low-carbon performance, and safe operation of the electric-thermal-hydrogen MES.

Key words: carbon emission reduction, carbon trading mechanism, hydrogen, integrated demand response, multi-carrier energy systems, multi-objective optimization, non-dominated sorting genetic algorithm II