Multi‐stage coordinated planning of energy stations and networks in park‐level integrated energy systems
Yue Qiu,Yunting Yao,Suyang Zhou,Hanlin Zhang,Yuqing Bao
DOI: https://doi.org/10.1049/rpg2.13166
IF: 3.034
2024-11-29
IET Renewable Power Generation
Abstract:This paper proposes a multi‐stage station‐network coordinated planning method for park‐level IES with the integration of distributed renewable energy sources. Considering the long‐term evolution of multi‐energy loads, energy prices, and technical parameters in the planning area, the multi‐stage planning method aims to achieve phased and orderly investments in energy stations, load‐side equipment, and electricity, gas, and heat/cold networks. A chance‐constrained programming method is used to model the uncertainty of distributed renewable energy output. Additionally, the improved big‐M method and the second‐order cone relaxation technique are adopted to process the multi‐stage planning model into a mixed‐integer second‐order cone programming form. With the development of distributed energy resources and intelligent energy management technologies, park‐level integrated energy systems (PIESs) are essential for multi‐energy flow conversion and utilization. However, coordinating the upstream energy sources, internal energy stations, distribution networks, and downstream loads within the PIES framework remains a challenge. Developing a station‐network coordinated planning scheme for PIES confronts challenges, such as overly complex models that are difficult to solve and the difficulty of implementing standard methods for medium and long‐term planning. This paper proposes a multi‐stage coordinated planning approach for PIES, containing energy stations, multi‐energy networks, and load aggregation nodes. The energy equipment and energy networks are precisely modelled to enhance the reliability of the planning scheme. The chance‐constrained programming (CCP) method is adopted to address uncertainties arising from renewable energy generation. Additionally, the improved big‐M method and second‐order cone (SOC) relaxation technique are utilized to manage non‐linear elements, transforming the model into a mixed‐integer second‐order cone programming (MISOCP) form to enhance solution efficiency. A case study illustrates that the proposed method enables effective allocation of equipment and networks, outperforming single‐stage planning by reducing investment costs, enhancing facility utilization rates, and facilitating renewable energy integration.
energy & fuels,engineering, electrical & electronic,green & sustainable science & technology