Studies On The Wind Speed And Wind Power Forecasting In Wind Farm

Wind power is the fastest growing renewable energy power generation, in the world. In recent years, wind power has been rapidly developed in China. But Wind power is volatility and randomness, which will bring challenge to the safety and stabilization of power grid and then restrict the scale of wind power development. Short-term wind power prediction is an effective approach for the above problem. Based on historical data, this paper studies the wind speed and power forecasting in the wind farm.Wind speed forecasting is the Foundation of wind power forecasting. This paper proposes a model based on time series and Support Vector Machine for wind power forecasting. The model needs to determine the input variables, the selection of training samples and parameters optimization. In the proposed method the mathematical model was built by time series method to obtain factors having significant impacts on the wind speed, and these factors were used to choose input variables of SVM. A method base on time series trace evolution was used to search similar samples to the forecasting point as the training samples of the SVM model. The Particle swarm optimization was adopted to select the most optimal parameters. The actual examples prove that the model has good performance.Modeling of wind turbine power curve is one important step of wind power forecasting. This paper proposes a power matrix model based on wind speed, wind direction. The power matrix considered the situation that the same wind speed and the different wind direction output different wind power. Compared with Modeling of wind turbine power curve based on wind speed, the power matrix has more precise description of the changes in wind power. Examples show that the wind power forecasting used the power matrix obtain better prediction results.In order to solve the problem that the forecasting accuracy is not high only by certainty point forecasting model and difficult to describe the volatility characteristics of wind power, the wind power forecasting from the point forecasting research expanded to include the uncertainty of wind power fluctuation interval forecasting. In this paper, V-support vector machine is employed to establish wind power fluctuation interval prediction model which does not require statistical analysis of errors and assume the probability distribution of forecast errors. According to decision-making requirements, adjust the model parameters output the corresponding confidence levelof the wind power interval. The actual examples prove that the model point forecast has better accuracy and fluctuation interval forecast can effectively characterize the wind power fluctuation, and its provides a basis for the risk decision-making in the power system dispatch and control.