Generation–grid–load–storage coordination for a park-level integrated energy microgrid under new-type power systems

Enhancing energy allocation efficiency in park-level Integrated Energy Microgrids (IEMs) is crucial for advancing New-Type Power Systems (NTPS). This study develops a Generation–Grid–Load–Storage (GGLS) collaborative coordination framework designed to address the operational challenges associated with high wind power penetration and its inherent stochasticity. A stochastic opportunity-constrained planning method is formulated to accurately characterize the variability of wind generation. To handle the large-scale uncertainty scenarios, a Backpropagation Neural Network (BPNN) is employed for dimensionality reduction and representative scenario extraction. Furthermore, a collaborative optimization model for the IEM is constructed by incorporating multiple energy forms, inter-temporal coupling relationships, and uncertainty-aware operational constraints. A hybrid solution strategy, integrating the Quantum-Inspired Evolutionary Algorithm (QIEA) and the Genetic Algorithm (GA), is proposed to efficiently solve this complex multi-objective model. Simulation analyses demonstrate that the proposed method notably improves economic performance, energy utilization, and operational robustness while mitigating the adverse effects of wind uncertainty. Overall, the proposed GGLS coordination framework offers a scalable and resilient pathway for future IEM planning aligned with the goals of NTPS. • Developed GGLS coordination for park-level IEMs under NTPSs. • Modeled stochastic wind uncertainty via opportunity-constrained planning. • Applied BP neural network for representative scenario reduction. • Proposed hybrid QIEA–GA for global–local collaborative optimization. • Improved cost efficiency and energy utilization under uncertainty.