中文摘要：

随着风电接入电网规模的日益扩大,其波动对电网的影响也日趋增大,而准确的风电功率预测可以有效地降低其对电力系统稳定性的影响,因此风电功率预测对接入大规模风电的电力系统稳定运行具有相当重要的意义。提出一种模糊粒计算和支持向量机相结合的风电功率实时预测方法,利用模糊粒计算将风电功率时间序列划分为简单的子序列时间窗口,同时把具有相似属性的对象组合在一起,通过提取核心信息减少冗余,利用支持向量机法对子序列进行预测,得到最终的预测值。以东北地区某两个风电场的实测数据为例,根据国家能源局文件中的指标验证了模型有效性。

英文摘要：

With the expansion of the wind power the fluctuation of the impact on the grid is increasing, and the accurate wind power forecasting can valid decrease its affect to power system stability, so the wind power prediction is very important to the stability of power system operation, when large-scale wind power access to power system. This paper proposed a kind of fuzzy granular computing and support vector machine united wind power forecasting method based on fuzzy granular computing, the wind power time series is divided into sub-sequences of simple time window, while the objects with similar attributes together, reduce redundancy by extracting core information, prediction using support vector machine sub sequence, to get the final prediction value. Taking the measured data of a two wind farm in Northeast China as an example, the validity of the model is verified on the basis the index of the National Energy Bureau.