Multi-objective evolutionary algorithm with two-tier fully-connected weight network for day-ahead scheduling of integrated cooling, heating and power energy systems

The day-ahead scheduling of an integrated energy system, which combines cooling, heating, and power generation alongside wind and photovoltaic energy sources, presents several challenges. These challenges stem from the presence of semi-continuous variables, multiple optimization objectives, nonlinearities, and various constraints. Existing multi-objective evolutionary algorithms, however, are not sufficiently efficient in solving the day-ahead scheduling model, particularly when dealing with semi-continuous variables. To address these issues, this paper proposed a novel multi-objective evolutionary algorithm based on a two-tier fully connected weight network. The model leverages a fully connected network to effectively address the semi-continuous variable problem, offering a unique solution to the complexities of day-ahead operation scheduling. The upper layer of the network employs multi-space dimensionality reduction to enhance global search capabilities, while the lower layer focuses on local search for more targeted solutions. Furthermore, the design of constraint repair operators is influenced by the structure of the algorithm, aiming to satisfy the constraints inherent in day-ahead scheduling and improve the efficiency of the search process. In experimental simulations, the proposed algorithm’s performance was compared with that of the latest constrained multi-objective optimization algorithms. The results demonstrate that the proposed algorithm significantly improves both optimization efficiency and the quality of the scheduling solutions. These findings highlight the effectiveness and superiority of the proposed algorithm in optimizing day-ahead scheduling for integrated energy systems.