Zonotope approximation based flexible cluster division method for load-side resource scheduling

Abstract

In view of the insufficient flexibility and adjustment capabilities caused by the high proportion of renewable energy connected to the distribution network and the limitations of the source-side adjustment method, a load-side resource flexibility cluster partitioning method based on zonotope approximation is proposed to cope with the flexibility adjustment requirements brought by uncertain resources. First, the constraint space of load-side resources such as energy storage, multi-source mobile energy storage, electric vehicles, and temperature-controlled loads is described as a convex polyhedral form general model based on the virtual battery framework, and the zonotope approximation is used to integrate its constraint space with the Minkowski sum. Secondly, modularity and flexibility supply and demand balance index are proposed to perform flexibility clustering of load-side resources. Then, the minimum comprehensive operating cost of the cluster aggregator is set as the optimization goal, the cluster constraint space is equivalent to the cost function, and the optimal charging and discharging power and flexibility adjustment capacity of the load-side resources are obtained. Finally, the method proposed here is verified through numerical example simulation to effectively balance aggregation accuracy and calculation speed.