摘要
将风电场理论功率和实际功率之差作为弃风电量的时间序列,利用其混沌性对其进行相空间重构,采用人工蜂群算法(ABC)优化的最小二乘支持向量机(LSSVM)参数,获取弃风电量预测模型(ABC-LSSVM)。首先,将弃风电量数据进行归一化处理,减小数据上下限之间的差距,提高预测模型的泛化能力;然后,将弃风电量时间序列进行相空间重构建立数据模型;最后,把数据模型带入预测模型中完成预测。本文以新疆达坂城某风电场数据为例,对基于人工蜂群算法的最小二乘支持向量机进行了仿真,结果表明,此方法能很好的预测出弃风电量的变化趋势,对弃风规划有一定的指导意义。
The difference between the theoretical power and the actual power of wind farm is taken as the time series of wind power,and the phase space is reconstructed by its chaotic property.The least squares support vector machine(LSSVM)optimized by the artificial bee colony algorithm(ABC)is used to built the abandoned wind power prediction model(ABC-LSSVM).Firstly,the abandoned wind power data is normalized to reduce the gap between the upper and lower limits of the data and improve the generalization ability of prediction model.Then the time series of abandoned wind power is reconstructed to establish the data model.Finally,the data model is input into the prediction model to complete the forecast simulation.Taking a wind farm data in Dabancheng,Xinjiang as an example,the ABC-LSSVM model is simulated.The results show that this method can predict the change trend of abandoned wind power better,and has a certain guiding significance for abandoned wind power planning.
作者
谢丽蓉
杨欢
轩武警
包洪印
XIE Lirong;YANG Huan;XUAN Wujing;BAO Hongyin(School of Electrical Engineering,Xinjiang University,Urumqi 830047,Xinjiang,China;School of Aerospace Engineering,Xiamen University,Xiamen 361005,Fujian,China;China Shipbuilding Heavy Industry Haiwei(Xinjiang)New Energy Co.,Ltd.,Urumqi 830002,Xinjiang,China)
出处
《水力发电》
北大核心
2019年第12期101-104,共4页
Water Power
基金
新疆维吾尔自治区区域协同创新专项(科技援疆计划)(2018E02072)
关键词
弃风电量
人工蜂群
最小二乘支持向量机
预测
abandoned wind power
artificial bee colony
least squares support vector machine
prediction
