Implementation of interior point method based voltage/reactive power optimization

In the real time voltage/reactive power optimal control, it is more and more important to expedite the calculation speed and improve robustness by dealing with the infeasibility problem, because the number of control is very limited and more constraints have to be considered in a large or weak-meshed power system. In this paper, a new voltage/reactive power optimization method based on primal-dual interior point method is presented. A fuzzy theory based model to handle the infeasibility problem is proposed. Compared with least square based infeasibility handling method, a carefully adjusted weight matrix is eliminated while the speed is higher. Infeasibility detection is processed with the information of dual variables and complementary gap. Several techniques are discussed to improve the efficiency of the whole optimization program. Test results in East-China power system demonstrate that the proposed method is very efficient in a large scale power system, and is very effective when infeasibility occurs during the optimization