Research On Economic Dispatch Of Power System With Large Scale Wind Power Integration

Energy is the foundation of economic and social development of a country.China has brought sustainable development into economic development goals,attached importance to environmental protection,put forward the great goal of "carbon reaching peak and carbon neutralization".It is striving to reach the peak in carbon dioxide emissions by 2030 and realize "carbon neutralization" by 2060 years.Therefore,wind power generation,as a renewable energy,has been widely used,with large reserves and wide distribution in the case of excellent wind resources distribution in China.However,with the gradual expansion of wind power scale,wind power also presents a new challenge to the operation of power system while providing advantages.Therefore,it is very important to establish a large-scale wind power generation system,a new and more accurate economic operation mode,to maintain the safe and stable operation of the power system and improve the economic efficiency under the condition of making full use of the new energy power.The specific work contents are as follows:(1)After learning the economic dispatching method of power system,and the principle and characteristics of BP neural network algorithm and ant colony algorithm,according to the needs of this project,the wind power prediction and the actual situation of power system economic dispatching are combined with the related characteristics of BP neural network algorithm and ant colony algorithm,and then the appropriate optimization scheme is selected.(2)Through the analysis of the influencing factors of wind power forecasting,the input data of wind power forecasting is determined and preprocessed.Then,genetic algorithm is used to improve BP neural network algorithm.The wind power generation prediction is carried out by using this method.The final wind energy interval prediction is output after the prediction error is obtained based on the non parameter estimation calculation.(3)A dynamic economic dispatching model of power system is established,which takes into account the fuel cost,environmental pollution control cost of thermal power unit and the cost of start and stop of thermal power units,as well as the cost of wind power generation and penalty of wind power.(4)The advantages and disadvantages of the basic ant colony algorithm are studied and analyzed.On the basis of maintaining its advantages,the shortcomings are improved by integrating other optimization algorithms,and then matlab programming is carried out.(5)The wind power prediction is carried out by using the improved BP neural network algorithm.Then,the IEEE-39 node is analyzed by combining the original ant colony algorithm and two improved ant colony algorithm.The results show that the new improved ant colony algorithm can save 2% more cost than the original ant colony algorithm in solving the economic scheduling problems of power system.Besides the economic improvement,it has the advantages of fast speed and excellent unit output ratio.