Data-Driven Volt-VAR Coordinated Scheduling With Mobile Energy Storage System for Active Distribution Network

Abstract

In order to improve the voltage distribution and operation cost for ADN, A scheduling strategy is designed to integrate flexible resources, particularly mobile energy storage systems, within the coupling of ADN and TN, under an uncertain environment. A day-ahead Volt-VAR coordinated scheduling framework for ADN and TN can be proposed through incorporating a data-driven day-ahead scenario generation method based on denoising diffusion probabilistic model. First, the historical data may be employed to learn the error relationship between real power curves and predicted power curves for generating RES scenarios. The probability distribution for prediction error is constructed which can describe the day-ahead output power curve of RES. Subsequently, a SCCO approach is employed to measure voltage operating risk in uncertain environment, which can effectively utilize the controllability of different resource at timescale and spatial scale in ADN to fulfill the anticipated operational requirement. Finally, The ADN coupled with TN model can be linearized and converted into the mixed-integer linear programming method. Numerical simulations based on the IEEE 33-bus distribution system coupled with 15-node transportation network may verify the effectiveness of the proposed method.