基于短期风功率预测的数据预处理算法研究  

Research on data preprocessing policy based on short-term wind power prediction

作　　者：许梦田 王洪哲 赵成萍[1] 严华[1] Xu Mengtian;Wang Hongzhe;Zhao Chengping;Yan Hua(School of Electronic Information,Sichuan University,Chengdu 610015,China;Power Dispatch and Control Center State Grid Liaoning Electric Power Supply Co.Ltd.,Shenyang 110006,China)

机构地区：[1]四川大学电子信息学院,四川成都610065 [2]国网辽宁省电力有限公司电力调度控制中心,辽宁沈阳110006

出　　处：《可再生能源》2019年第1期119-125,共7页Renewable Energy Resources

基　　金：国家自然科学基金项目(61172181)

摘　　要：针对实测风速和功率数据中包含奇异点以及同一风速下风功率存在较大范围波动的问题,文章提出一种数据预处理算法。首先,采用拉依达准则剔除风速和功率奇异点;再使用优化的一次指数平滑法及最大皮尔逊相关系数对风速进行平滑处理;最后,利用新疆阿勒泰地区某风电场单台风机的实测数据进行验证分析。以文章提出的预处理方法得到的风速作为BP神经网络预测模型的输入,风功率的预测准确度显著高于已有预处理方法得到的结果。Aiming at the problem that the measured wind speed and power data contain singular points and there exists fluctuations in wind power at the same wind speed,a data preprocessing algorithm is proposed.Firstly,the Pauta criterion is adopted to eliminate wind speed and power singularities.Then,the improved single exponential smoothing and maximum Pearson correlation coefficient are used to smooth wind speeds.Finally,the measured data of a single wind turbine of a wind farm in the Altay region of Xinjiang were used for verification and analysis.The wind speed obtained by the pretreatment method proposed in this paper is taken as the input of BP neural network prediction model,and the prediction accuracy of wind power is significantly higher than that obtained by the existing pretreatment method.

关 键 词：数据预处理 拉依达准则 一次指数平滑 皮尔逊相关系数 风功率预测 

分 类 号：TK81[动力工程及工程热物理—流体机械及工程]