Research On Optimal Dispatch Of Power System Considering Large-scale Wind Power

With the rapid growth of wind power, more wind power plants have been connected to the power grid.However, the randomness, volatility and the anti-peaking characteristics of wind power bring forward new challenges to the power system analysis, scheduling and control. To solve this problem, power system dispatching management based on wind power prediction technique is the most economic and effective way. The major work of this paper includes:Proposes a method based on long memory time series model which is used to predict short-term wind farm power. Considering the shortage about traditional model, a long memory model based on ARFIMA has been put forward. The model is used to predict wind farm power in this paper Due to the problem that there is a missing and the wrong about historical data, EN algorithm is put forward to deal with incomplete data about wind farm power. Analyzing by predicting results, the short-term prediction of wind farm power which is based on long memory model improves the accuracy of wind farm power prediction. It has a practical value.Proposes three scheduling optimization models which contain generation cost?environmental costs and risk costs caused by grid containing wind power. The probability density of load forecasting value and wind power forecasting value are used to predict loss of load expectations and abandoned wind expectations. And then the expectations are introduced into the model as risk costs. Optimal scheduling result can be obtained by the relationship between contain generation cost?environmental costs and risk costs. It ensures the safety and economy of power systems effectively. When selecting scheduling scheme, generation cost ?environmental costs and risk costs are weighted by their own importance, and the new scheduling scheme can meet the different policies and regulations.Due to the shortage of particle swarm optimization, an improved particle swarm optimization based on estimation distribution has been put forward. In this method,the probability of unit commitment will be generated by unit capacity and the capacity range of parameters, and then multi-unit operation status will be put forward. By using estimation distribution algorithms, the problem about low search efficiency of traditional particle swarm optimization will be improved. At last, the optimization results about system units and load distribution will be obtained by improved particle swarm optimization. Simulation results show that wind power can significantly reduce total system cost and pollution gas emissions. It also proves that the particle swarm optimization based on estimation distribution has a use value which possesses high accuracy and convergence rate.