Low-Carbon Robust Planning for PIESs with Multi-Time-Scale Uncertainties and Elastic DR Regulation

With the widespread application of park integrated energy systems (PIESs), challenges of multi-energy coupling, high investment costs, and multi-type uncertainties have become increasingly prominent. Existing research often employs typical scenario generation or robust optimization for short-term uncertainties but struggles with long-term load growth uncertainties and fails to fully utilize the flexibility of demand-side resources during the planning phase. This paper proposes a robust planning method for PIESs considering dynamic demand response and multi-timescale uncertainties. First, an energy flow framework encompassing cooling, heating, electricity, gas, and hydrogen is constructed. To overcome the limitations of traditional fixed-boundary DR, a dynamic elastic DR mechanism featuring transferable, substitutable, and curtailable types is established. Transferable demand boundaries are defined by a price–demand elasticity matrix, and actual responses are dynamically adjusted in synergy with system power balance conditions for optimal configuration. Second, multivariate dynamic time warping and hierarchical clustering algorithms derive typical daily scenarios accounting for short-term uncertainties. Finally, information gap decision theory characterizes long-term load growth uncertainty, constructing a robust planning model addressing both timescales. Case studies show that flexible resources and demand response reduce lifecycle cost by 55.24% and carbon emissions by 47.75%. The proposed demand response method further cuts costs by 153,800 yuan and emissions by 11.36%. The robust planning method synergistically addresses multi-timescale uncertainties, ensuring economy while maximizing resilience to uncertain fluctuations.