Research Of The Wind Speed And Power Generation Probability Forecasting In Wind Farm

With the increasing severity of energy and environment problems, renewable energy is competitively developed all over the world. Wind energy, which is clean and abundant, is attached great importance to and becomes the fastest developing renewable energy. As the rating of wind power in power network rises, the large wind farms often reach several hundred MW. Unlike the other power plants, the active power from wind farms varies all the time, which may influence the operation and stability of power network after connection. As a result, the impact of wind power on power system should be analyzed in detail. The research on wind power is deeply and widely at home and abroad. But wind speed forecasting and wind power forecasting in wind farm still can't make us satisfied. And in our country few people work on this subject. This paper makes research and discussion in the field.Firstly, the paper introduces the research status of the topic at home and abroad, And clarify the sequence of time, neural network and other major existing method of the basic principles in this area of application. Secondly, after contrasting the above in many ways, this paper introduces a new method - Least Squares Support Vector Machine method to forecast the wind speed. It establishes the prediction model by using the above method with the platform of Matlab toolbox LS-SVM lab. After using the rolling weights, the network reflect the change of the wind speed. Test show that the model has some practical value.This paper introduced the v-Support Vector Machine method, established the probability of the wind speed prediction model. It provides the wind speed probability forecast and brings a wind speed range with the degree of confidence. And without prior forecast deviation of the distribution of the probability of assumptions, it can give more credibility at the same time the confidence interval estimation. In the current wind speed is not high prediction accuracy of the circumstances, wind speed prediction of the probability of providing a new way of thinking.Finally, the paper obtains equivalent single wind farm output from the wind power curve, and compares the accuracy between the wind speed forecasting and the wind power generation , pointing out that because of wind speed and wind power generation different from the strength of their own laws, leading to the prediction accuracy of their corresponding gap also exists.