Research On Short-Term Prediction Models Of Wind Speed And Wind Power

With the rapid development of wind power in the world, the proportion of wind power in power grid is larger and larger. However, the development of wind power is limited due to its intermittence and randomicity when large-scale wind energy access to the grid. Therefore, accurate prediction of the output of the wind power is necessary so as to achieve the optimal operation and dispatching of the power system, as well as reducing the power system spinning reserve and operating costs.Under this background, the short-term wind speed and power prediction was researched in this thesis. Various methods were used to study on short-term wind speed and power prediction. Firstly, time series and neural network were introduced into wind speed prediction modes, and ARIMA and BP neural network prediction model were established respectively. Then two improved method of GARCH model and clustering ARIMA model were proposed based on ARIMA model. And genetic algorithm and particle swarm optimization were introduced into BP neural network, so GA-BP and PSO-BP prediction models were established to optimize the initial weights and thresholds of BP neural network. Finally, the prediction results and feature of each model were analyzed through examples.In order to improve the prediction accuracy, the combination model was introduced into wind speed prediction. Combination prediction result of wind speed was obtained through combining each single model's prediction results. Finally, combination prediction result of wind speed was converted into wind power prediction results based on wind power curve.From what has been discussed above, the prediction models were discussed in-depth by using time series and neural network theory, and data processing and numerical calculation was carried out. So, it can be concluded that the wind speed and power prediction results can be improved through model modifications and data processing.