Research On Wind Speed Prediction System Of Wind Farm Based On Mixed Programming Of VC++and MATLAB

With the emergence of the global energy crisis, the world will turn their attention to renewable energy. Wind energy is a very important clean energy, it is safety, environmentally friendly. It can provide stable, long-term supply of resources. Wind power is the main form of wind, and is currently the most mature, most large-scale development conditions and prospects for commercial development.Wind speed forecast has been a hot research. If the forecasts for wind farms are more accurate, it will help to adjust scheduling, to improve wind power penetration limit, to reduce adverse effects on the entire grid effectively, to reduce operating costs and spinning reserve of the power system, etc. This you can work out the correct energy exchange program in an open electricity market environment.In this paper, we combine the theory and practice base on SVR. Main work is as follows: First, we use the powerful data processing capabilities of MATLAB software to achieve SVR to predict the wind speed. Then we study the modeling of SVR in-depth.As the randomness of wind speed, feature extraction has some difficulties, but also features selection is also very subjective, easily lead to insufficient information or redundant. In order to improve the accuracy of wind speed forecast the wind speed time series forecasting method of SVR based on phase space reconstruction is proposed. First, we reconstruct the phase space of the time sequence, and then use the phase point of reconstructed phase space as input of features, and train the model. This method can be expected according to the internal laws of the time series and constructing the input characteristics adaptively. It makes the prediction accuracy and the length of the time of predict effectively increased substantially.Finally, we use VC++to write visualization software of the method. In order to reduce development effort and shorten the development cycle, full MATLAB and VC++language to their strengths, we implement complex algorithms in MATLAB and write interface in VC++. To achieve the process of program development efficiently and the speed of program execution effectively through MATLAB and VC++mixed programming. You can always use the software to add, modify, query wind data. The data are stored in the database directly. Ultimately wind speed is forecasted in the software platform.This paper discusses the significance of wind speed prediction and the research status based on literature research. In the analysis of forecasting methods of traditional, a combination of SVR and phase space reconstruction is applied to wind speed prediction. Trying to establish a new method of wind speed forecast of SVR based on phase space reconstruction. This method can be expected according to the internal laws of the time series and constructing the input characteristics adaptively. It makes the prediction accuracy and the length of the time of predict effectively increased substantially. And using a powerful programming language:VC++and MATLAB as a platform to write a platform of wind speed prediction software that helps to promote the progress of wind speed forecast.