Design And Implementation Of Short-term ZLNN Wind Farm Power Prediction System

In recent years, because of China’s ecological environment more and more serious problem as well as the country to vigorously support the strategy of sustainable development, wind power as a non polluting and renewable resources has been the rapid development. But because the wind is fluctuating and strong randomness, and with the power of wind power exceeds a certain value, it will wind power network scale a strong impact on the power system, and caused great impact to power transmission efficiency and quality. The significance of accurate wind power forecasting is very important to improve the stability of wind power system exists, but also can improve the wind power price effective competition in the power system, but also can bring certain convenience for wind power equipment maintenance personnel, so the wind power to make an accurate prediction of the more the necessity of becoming. The research of this field are more and more widely and deeply, foreign countries have multiple sets of prediction system of mature into the actual operation, but the domestic research in this field has not yet reached a satisfactory level, nor the use of high accuracy of the system is put into use. The National Energy Bureau in June 22, 2011 issued a "wind power prediction notification" Interim Measures of forecast management, requirement each wind farm must be in accordance with the notice requirements of the implementation of the installation of wind power forecasting system. Again in February 27, 2012 issued a "wind power forecast coordination with the grid operation implementation rules(Trial)", on the power of the forecast more detailed constraints and rules.ZLNN is engaged in large-scale wind farm management and operation of energy companies. Short term power ZLNN wind farm in this paper to improve the forecasting system of wind power forecasting based on full utilization, improve the power grid scheduling and resource, implementation develops a set to conform to the wind farm side oriented application of grid scheduling system software specification and safety requirements of the power prediction system platform. Through computer simulation, to study the forecast model and method, advantages and disadvantages and applicable scope of various methods of analysis, and the effect of various models and methods of comprehensive use; puts forward 2 kinds of suitable model, provides the technical foundation for the development of prediction system. Based on study of prediction model and method, choosing forecasting model and method of suitable, development oriented wind farm side application power prediction system.The system itself has the function of wind power integrated information communication, can simultaneously run with a variety of wind power monitoring system, to realize the seamless data read. Prediction system software is developed by Java/Python language, using Web based on B/S architecture, is a cross running operating system platform. Switching software structure and data specification of power dispatching technical support system specifications, can be very good and the power grid dispatching system with power system two times and meet the safety protection requirements, has the good man-machine interface. The system can provide 0-4 hours of super short-term power prediction and 24 hours of short-term power prediction, and can be the result of forecasting uncertainty are analyzed and estimated.