Research on probabilistic reactive power optimization considering the randomness of distribution network

With the rapid development of intelligent distribution network, the uncertainty of the load and the randomness of distributed generation have brought new challenges to distribution network control operation especial in reactive power optimization. This paper uses probabilistic power flow algorithm based on three-point estimate method to solve the uncertainty caused by power flow calculation in the stochastic models of load and wind power so as to propose a method of information entropy principle to measure the voltage fluctuation. On the basis of this method, a model of probabilistic reactive power optimization considering minimum network loss and voltage fluctuation is built. Taking the IEEE 33 nodes system which contains wind power generation as an example and we draw a conclusion that if we add the minimum voltage entropy to multi-objective reactive power optimization objective function, the probability distribution of node voltage is more centralized than that of single objective reactive power optimization. Thus, to optimize reactive power by means of this model could improve the voltage stability of the system and make the voltage distribution near a certain value that within the scope of control in large probability. The proposed multi-objective probabilistic reactive power optimization model is suitable for the actual distribution network reactive voltage control with random properties.