A two-layer optimal configuration method for integrated energy systems based on energy cascade utilization: structure-capacity-operation co-optimization

Global calls for clean energy and carbon neutrality have drawn significant attention to integrated energy systems (IESs). The optimal configuration of IESs, which determines the types, interconnections, and capacities of energy conversion and storage components, is essential for achieving efficient, economical, and low-carbon operation. Traditional configuration methods often rely on predefined rules (e.g., equipment interconnection sequences and operation priorities) to simplify optimization. However, these rules restrict configuration flexibility and hinder the achievement of optimal solutions. Therefore, this study proposes a two-layer optimal configuration method for IESs that eliminates rule-based constraints, enabling co-optimization of equipment selection, interconnection, capacity, and operation scheme. Firstly, a layering method of IESs is proposed based on the principle of energy cascade utilization, enhancing the flexibility of equipment configuration combinations. Secondly, a two-layer constrained optimization evolutionary framework is developed. The upper layer optimizes equipment selection and interconnection relationships, considering both economic and exergy indicators. The lower layer formulates constraints based on the upper-layer structure scheme, and jointly optimizes the capacities and operation schemes of renewable energy, energy conversion and storage equipment. The lower-layer results are iteratively fed back to the upper layer until convergence is achieved. Finally, five comparative cases are established to verify the feasibility and effectiveness of the proposed method. Simulation results show that the proposed method has substantial advantages over the optimal configuration based on a pre-specified IES structure. The values of annual total cost saving rate and exergy saving rate increase by 5.96% and 7.44%, respectively. Furthermore, the sensitivity analysis demonstrates that the proposed method exhibits favorable robustness.