Research On Dynamic Economic Dispatch Of Wind Power Intergrated Systems

Wind power as a clean pollution-free and renewable energy get the attention of the countries all over the world more than before, generation cost of wind power is low and dose not cause pollution that have advantages compared with thermal unit, but wind affected by many kinds of natural factors have strong randomness, intermittent, and volatility, the predictive difficulty of wind speed and wind power is much larger than the load, the uncertainty of wind power integrated systems has influenced the safety and economic operation of the system, through configuring appropriate spinning reserve to cope with the volatility and risk of system, wind power integrated into system brings with challenges to the electric power system, especially economic dispatch, the unit commitment, which is the core of the economic dispatch problem facing the severe test.On the bases of domestic and foreign research results and the understanding of problems about unit commitment, putting forward through configuring spinning is the effect of wind prediction, adopting the method of cost-benefit analysis, Through the mutual check of the spare benefit and cost in the objective function, suitable spare could be automatic configured for the system, and part of the model that contains{0,1} variable product form of nonlinear constrains would be transformed to a group of linear constrains to express, the transformation of the mathematical model was a standard secondary mixed integer programming problem.Firstly the model and solving method of economic dispatch is introduced, wind speed and wind power model and dynamic economic dispatch model of wind power integrated system are analyzed, the basic principle of particle swarm optimization(PSO) is also introduced, and the process of solving dynamic economic dispatch problem using PSO is recommended.Secondly the idea of adding simulated annealing into particle swarm algorithm is applied to deal with the algorithm disadvantage of falling into local optimum, cooperate with outside point method processing in violation of the constraint condition, the influence factors of dynamic economic of wind power integrated systems is analyzed, unit shutdown, the impact of wind power and load forecasting error are considered and spinning reserve is introduced to cope with the possible power shortage and its cost to join to the objective function and the cutting load probability is regarded as the risk index, under the assumption that wind speed and load forecasting error under the premise of accord with normal distribution, the probability constraints expression is converted into a certain inequality constraints, dynamic economic dispatch model is solved based on the idea of simulated annealing idea of hybrid particle swarm algorithm.Finally, a practical combined linear mixed integer model and solving method of unit commitment, the form of nonlinear constrains would be transformed to a group of linear constrains to express, the transformation of the mathematical model was a standard secondary mixed integer programming problem, it could solved by MIP commercial software.The reserve configuration is an important aspect of the unit commitment problem and could affect system stability, the more reserve the more stable the system could be, but the corresponding cost would be more; the less reserve the less cost could be, but the corresponding losing load probability and wind power would be more, the risk of system could increase, this paper puts forward a model and solving method under certain reserve to deal with the uncertainty of load and wind power considering the problem of spinning reserve configuration, at last, a model based on a comprehensive method of cost-benefit analysis is put forward to processing the uncertainty of load and wind power, through the mutual check of the spare benefit and cost in the objective function, suitable reserve could be automatic configured for the system, artificial factors are avoided and could be simplified through mathematical method, which is easy to be solved by linear mixed integer programming model.