An innovative stochastic fuzzy-optimized dispatch strategy for multi-energy parks with uncertain opportunity constraints

Amid the wave of green and low-carbon energy transition, unprecedented acceleration is urgently required for global renewable energy deployment. However, complex interdependencies are created by the interplay between the random nature of renewable power generation and diversified energy demands, and the scheduling robustness of Regional Integrated Energy Systems (RIES) is undermined by these interdependencies. A Regional Integrated Energy System solution based on a fuzzy adaptive scheduling approach is proposed in this paper. Energy flexibility is maximized through the implementation of multi-domain collaborative optimization to dynamically balance supply–demand uncertainties. Firstly, a fuzzy probabilistic constraint programming approach is proposed, in which wind power, photovoltaic power generation, and load are treated as fuzzy variables, and a credibility measure is introduced to mitigate decision ambiguity. Secondly, novel fuzzy membership functions are designed to comprehensively characterize the uncertainty in renewable energy generation and electricity consumption. Thirdly, robust flexibility coordination for bidirectional source-load matching is achieved through a fuzzy adaptive mechanism, with combined membership functions enhancing optimization reliability. Finally, fuzzy constraints are converted into deterministic equations via an exact equivalence class solver, and the confidence level of the time step is optimized by an improved particle swarm optimization (IPSO) algorithm—characterized by a linear decreasing inertia weight based on the arctangent function. Research findings indicate that dispatch costs are significantly increased by the uncertainty in power supply loads (costs under multi-source uncertainty scenarios are 51.2% higher than those under deterministic scenarios), while a confidence level of 0.7 is critical for balancing system reliability and economic efficiency.