Search

Strong interdependence between power production and heat supply from combined power and heat units could restrict the wind power integration. Deploying hydrogen storage, typically formed by the electrolyser, hydrogen tank, and fuel cell, could be a promising measure to accommodate surplus wind power and facilitate the coordinated operation of power and heat. The authors consider coordinated scheduling under wind power uncertainty for an electricity-heat coupled system with hydrogen storage and propose a distributionally robust joint chance-constrained coordinated scheduling (DRJCC-CS) model, where the waste heat from the electrolyser and fuel cell is incorporated. The authors design distributionally robust joint chance constraints to account for wind power uncertainty related constraints and derive their tractable inner and outer mixed-integer convex approximations. Consequently, the proposed DRJCC-CS model is casted into a mixed-integer convex programme. Case studies are conducted to demonstrate the effectiveness of the proposed DRJCC-CS method.

Integrated Hydrogen-Electrical (IHE) microgrids are desirable testbeds for the practice of carbon-neutral energy supply. This paper studies the IHE microgrids planning (IHEMP) under a dynamic perspective. To keep track with the fast development of hydrogen industry, we propose a multistage stochastic mixed-integer program (MS-MIP) formulation. It comprehensively considers the siting and sizing decisions of IHE microgrids, the dynamic expansion of distributed energy facilities, and the detailed operational model to derive a robust, flexible and profitable investment policy. Moreover, a scenario-tree based sampling strategy is leveraged to capture both the large-scale strategic uncertainties (e.g., the long-term growth of electric loads and hydrogen refueling demands, as well as the cost changes of system components) and fine-scale operating uncertainties (e.g., random variation of renewable energy outputs and loads) under different time scales. As the resulting formulation could be computationally very challenging, we develop a nested decomposition algorithm based on Stochastic Dual Dynamic Integer Programming (SDDiP). Case studies on exemplary IHE microgrids validate the effectiveness of our dynamic planning approach. Also, the customized SDDiP algorithm shows a superior solution capacity to handle large-scale MS-MIPs than the state-of-the-art solver (i.e., Gurobi) and a popular scenario-oriented decomposition method (i.e., progressive hedging algorithm).