Nested long-short-term temporal compression optimization for high-renewable integrated energy systems with hydrogen storage and heat recovery

Integrated energy systems with high-penetration renewables are essential for achieving low-carbon, high-efficiency energy systems. However, the seasonal supply-demand imbalance on long-term scales remains a significant challenge. Hydrogen energy storage offers a promising solution, but its conventional power-hydrogen-power storage mode faces challenges, including low overall efficiency and high-dimensional solution. To address this, this paper proposes the integration of waste heat recovery technology with hydrogen energy storage to enhance its efficiency. Additionally, a nested long-short-term temporal compression optimization method is introduced. Specifically, Latin Hypercube sampling, k-means clustering, and the elbow criterion are used to capture and extract the monthly fluctuation characteristics of net load. The clustering results are then used to construct a temporal compression matrix, which is applied to specific devices to reduce decision variables. This optimization approach is tested through a case study to verify its reliability. The results show that the proposed method effectively addresses the dual challenge of optimizing over both long-term scales and high-time resolution. In specific scenarios, it achieves a 45.03 % improvement in solving efficiency compared to traditional full-scale optimization. Compared to the reference time horizon compression method, the flexible supply capacity of devices with temporal compression is improved, this results in a 8.85 % reduction in solution error. Furthermore, the incorporation of waste heat recovery technology reduces the total cost by 1.16 % in annual operation.