The Wind Power Forecast System Base On BP&WRF Model

As a kind of abundant,low-cost,clean and renewable energy,wind power is increasingly brought to the attention of the countries all over the world in recent years.However,wind resource itself inherent characteristics of random,intermittent and controllability,and greatly restrict the large-scale development of wind power and use.On analyzing and summarizing domestic and foreign relevant wind power forecast on the basis of a variety of practical cases,joint localization of wind power forecast characteristics,design and implement a set of based on BP neural network and the WRF model of interactive view of forecast and analysis tools,to the wind power output of a timely and accurate prediction can improve the acceptance of wind power capacity,ease the large capacity of wind power to the local area for the stable operation of the power grid brings severe challenges.This paper discusses the design and implementation of wind power prediction system in the process of wind power prediction models at home and abroad were analyzed and summarized in this paper the development and present situation and key technology in practical application;In wind power prediction based on the analysis of the business requirements of the overall framework of design for wind power prediction system and relevant business model;Put forward the development of wind power prediction system of the overall solution and the technical route,and describes in detail the realization process of wind power prediction model,and finally achieve the practical application effect,finally gives the conclusion and prospect.Results of thesis is that wind power prediction system and medium-term power prediction model and super short-term power applied to the numerical weather model calculation,and the key technologies such as artificial neural network,which not only include the mesoscale WRF model,data assimilation,wind speed correction short-term limited area numerical weather forecast model of application;Short-term forecast model also thorough application of the BP neural network and data mining algorithms such as curve fitting,and the two prediction model is integrated into the same business forecast system,made into a separate general prediction data generation engine,to satisfy the business forecast model in dealing with flexible and varied on the business needs of extensibility,it not only can carry out quantitative forecast wind speed,also can undertake a variety of quantitative prediction of meteorological elements,well adapted to carry out the localization of wind power forecast,as well as the influence of wind power of various meteorological elements of the objective quantitative analysis of the subject objectives.Its comprehensive architectural design and technical implementation,is the BP neural network and the mesoscale WRF model comprehensive application in the field of wind power forecast a good attempt.