A wind power interval forecasting method based on Bayesian network forecasting model

The invention discloses a wind power interval prediction method of a Bayesian network prediction model. Based on the predicted wind speed and actual power data obtained by preprocessing, the relativity of Bayesian network is established, the range of variation of the interval variable amplitude parameters beta High and beta Low are obtained by calculation and comparison, genetic algorithm is usedto optimize the fitness value to obtain the optimal interval variation amplitudes beta High_best and beta High_best, and substituting the amplitudes into the Bayesian network model, substituting the test data to obtain the wind power prediction interval, and evalutating the prediction results by various evaluation indicators. It can not only get the fluctuation range of wind power, but also evaluate its reliability effectively, and provide effective reference for power dispatching decision-making. 本发明公开了种贝叶斯网络预测模型的风电功率区间预测方法，根据经过预处理得到的预测风速和实际功率数据，建立具有相关性的贝叶斯网络关系，计算比较得到区间变化幅值参数β和β的变化范围，运用遗传优化算法通过优化适应度值得到最优区间变化幅值β和β，带回贝