Coordinated Control of Active and Reactive Power of Distribution Network with Distributed Photovoltaic Based On Scene Analysis

Variables such as distributed generation (DG) output and load demand have randomness and volatility, brings large challenges to operation and control of distribution network. In order to solve this problem, this paper generates multiple scene samples through Latin hypercube sampling (LHS) based on mathematical distribution model of variables, then reduces the scene samples by K-means clustering, typical scenes and the probability of each typical scene are obtained. Considering the adjustable capabilities of distributed photovoltaic (PV) and energy storage system (ESS), a multi-objective optimization model based on scene analysis for coordinated control of active and reactive power in distribution network is established, and the mixed integer nonlinear programming model is solved by second-order cone relaxation (SOCR) technique. Finally, the improved IEEE 33-node system is taken as an example for simulation with results confirming accuracy and rationality of the proposed algorithm.