Research On Multi-objective Optimal Active Power Dispatch Considering Constraints Of N-1 Security And Pumped Storage Units’ Operation

Since the global energy situation and environmental problems are becoming more and more severe, it is of great concern to realize energy-saving and emission-reduction of electric power industry. In order to achieve the goal of energy-saving and emission-reduction, eco-nomic dispatch which simply minimizes the power generation cost as an objective can not meet the requirement of power grid operation, to develop power generation scheduling needs to consider multiple objectives such as power generation cost, emission of atmospheric pollu-tants and power purchase cost as well as intermittence of new energy resources, power grid security and other factors. A multi-objective optimal active power dispatch issue of power systems including wind farms and pumped storage stations which considers security, econo-my and environmental protection of generation schedule is studied in this paper. Main re-search and achievements is listed as follows.A multi-objective optimal active power dispatch model considering constraints of N-1 security and pumped storage units’ operation is proposed. Fuel consumption, emission of at-mospheric pollutants and power purchase cost are taken as objectives, the network security constraints of basic status and N-1 outage and operation dispatch of pumped storage stations are considered. In this model, DC load flow model is adopted to consider the constraints of network security, and based on which the effect of network loss is considered, so the power balance equation is more accurate. The up/down spinning reserve constraints are introduced to deal with the randomness of wind power which has an effect on power dispatch, two contin-uous variables(namely generating power and pumping power) which satisfy complementary condition, are introduced to describe the operation of pumped storage stations to avoid intro-duction of discrete variables, and the inequality constraints are reduced by means of accumu-lating operation mode constraints of a pumped storage station in all periods, also, the idea of “checking-adding-checking again-adding again”(Hereinafter referred to as N-1 check-and-add method) is used to handle network security constraints and hence the computational efficiency is enhanced. And comparison of the computational results of N-1 check and add method and the model with all the N-1 network security constraints are made to prove the correctness and efficiency of N-1 check and add method.The normal boundary intersection(NBI) algorithm, interior point method and N-1 check- and-adding method are adopted to solve this model. Firstly, NBI method is introduced to transform this multi-objective problem into a series of single objective optimization problems, and hence a series of uniform distributed Pareto optimal sets are obtained by means of the prime-dual interior point and N-1 check and add method. Secondly, entropy-weighted double base points method is applied to get a comprehensive optimal solution to provide a guidance decision for operators. Finally, parallel computing is applied to handle some sections of the algorithm, which can be computed concurrently, and the computational speed and practicality of the algorithm are improved. Test results on a real provincial power system with 167 units and 2299 nodes demonstrate that the proposed method can comprehensively consider the op-timal compromise among security, economy and environmental protection of generation schedule, and can develop the functions of pumped storage stations in shifting load and cop-ing with randomness of wind farm output and computing speed can meet the demand of prac-tical project applications.