| As one of the main ways of new energy power generation,wind power generation has the advantages of short construction period,low technical requirements,fast cost recovery,and less pollution,which is strongly advocated by various countries.Wind power forecasting is crucial to the operation and management of wind power plants.Accurate power forecasting results can help wind power plants make daily operational decisions,such as how to schedule the start-up and shutdown of generator sets,and how to perform load scheduling.In order to ensure the quality of wind power and reduce the impact of wind power grid-connected on the grid,there are also high requirements for the accuracy of wind power prediction.This paper first introduces the research status of many scholars at home and abroad on wind power prediction,finds the source of data sets and obtains data sets,and then uses Pearson correlation analysis method to analyze the correlation between various factors and wind power,finally determines the use of wind power according to the correlation.The traditional backpropagation(BP)neural network can be used as a prediction model,but it has the shortcoming of low accuracy.So as to improve the accuracy of the wind power forecasting model,this paper proposes a wind power forecasting model that optimizes the Seagull algorithm and improves the BP neural network.This paper also introduces the Logistic chaotic map into the basic seagull algorithm,which can make the individual distribution more uniform in the initial seagull population.By randomly perturbing the position of the initial seagull population,the algorithm can be prevented from falling into a local optimal solution.Changing the movement mode of individual seagulls in the seagull algorithm from linear movement to nonlinear movement,so as to improve the convergence ability of the algorithm.By using the improved seagull algorithm to find the optimal fitness individual as the initial weight and threshold of the BP neural network to establish a prediction model can greatly improve the performance of the BP neural network.This above model of the wind power prediction improves the accuracy of wind power forecasting when applying in Daqing area.This paper also compares the results of the proposed optimized the BP neural network model by improved seagull algorithm with other wind power forecasting models.The results show that the model proposed in this paper can effectively improve the accuracy of wind power forecasting and reduce the volatility of forecasting errors. |
