基于Adaboost的BP神经网络改进算法在短期风速预测中的应用Application of Adaboost-Based BP Neural Network for Short-Term Wind Speed Forecast
吴俊利;张步涵;王魁;
摘要(Abstract):
进行较准确的风速预测对含大规模风电场的电力系统进行经济调度具有重要意义。针对目前神经网络法、时间序列法、卡尔曼滤波法等算法在短期风速预测上精度不高的缺陷,引入Adaboost算法对前馈(back propagation,BP)神经网络算法进行改进,提出了基于Adaboost的BP神经网络算法,并将该方法应用于短期风速预测。经算例分析,该算法在超前1 h和2 h的风速预测精度优于其他2种算法,且该算法在高风速段(10 m/s以上)平均绝对百分比误差低于7.5%,具有较高的工程应用价值。
关键词(KeyWords): 风速预测;Adaboost;BP神经网络
基金项目(Foundation): 国家重点基础研究发展计划项目(973项目)(2010CB227206);; 国家863高技术基金项目(2011AA05A101)~~
作者(Author): 吴俊利;张步涵;王魁;
Email:
DOI: 10.13335/j.1000-3673.pst.2012.09.037
参考文献(References):
- [1]张宏宇,印永华,申洪,等.大规模风电接入后的系统调峰充裕性评估[J].中国电机工程学报,2011,31(22):26-31.Zhang Hongyu,Yin Yonghua,Shen Hong,et al.Peak-load regulatingadequacy evaluation associated with large-scale wind powerintegration[J].Proceedings of the CSEE,2011,31(22):26-31(inChinese).
- [2]张丽英,叶廷路,辛耀中,等.大规模风电接入电网的相关问题及措施[J].中国电机工程学报,2010,30(25):1-7.Zhang Liying,Ye Tinglu,Xin Yaozhong,et al.Problems and measuresof power grid accommodating large scale wind power[J].Proceedingsof the CSEE,2010,30(25):1-7(in Chinese).
- [3]吴俊,李国杰,孙元章.基于随机规划的并网风电场最大注入功率计算[J].电网技术,2007,31(14):15-19.Wu Jun,Li Guojie,Sun Yuanzhang.Maximum injection powercalculation of wind farms connected to power systems based onstochastic programming[J].Power System Technology,2007,31(14):15-19(in Chinese).
- [4]肖永山,王维庆,霍晓萍.基于神经网络的风电场风速时间序列研究预测[J].节能技术,2007,25(12):106-108.Xiao Yongshan,Wang Weiqing,Huo Xiaoping.Study on thetime-series wind speed forecasting of the wind farm based on neuralnetworks[J].Energy Conservation Technology,2007,25(12):106-108(in Chinese).
- [5]范高峰,王伟胜,刘纯,等.基于人工神经网络的风电功率预测[J].中国电机工程学报,2008,28(34):118-123.Fan Gaofeng,Wang Weisheng,Liu Chun,et al.Wind powerprediction based on artificial neural network[J].Proceedings of theCSEE,2008,28(34):118-123(in Chinese).
- [6]Li Shuhui,Wunsch D C,Giesselmann M G,et al.Using neuralnetworks to estimate wind turbine power generation[J].IEEE Trans onEnergy Conversion,2011,6(3):276-282.
- [7]Bossanyi E A.Short-term wind prediction using Kalman filters[J].Wind Engineering,1985,9(1):1-8.
- [8]马静波,杨洪耕.自适应卡尔曼滤波在店里短期负荷预测中的应用[J].电网技术,2005,29(1):75-79.Ma Jingbo,Yang Honggeng.Application of adaptive Kalman filter inpower system short-time load forecasting[J].Power SystemTechnology,2005,29(1):75-79(in Chinese).
- [9]潘迪夫,刘辉,李燕飞.基于时间序列分析和卡尔曼滤波算法的风电场风速预测优化模型[J].电网技术,2008,32(7):82-86.Pan Difu,LiuHui,Li Yanfei.A wind speed forecasting optimizationmodel for wind farms based on time series analysis and kalman filteralgorithm[J].Power System Technology,2008,32(7):82-86(inChinese).
- [10]Kamal L,Jafri Y Z.Time series models to simulate and forecasthourly averaged wind speed in Wuetta Pakistan[J].Solar Energy,1997,61(1):23-32.
- [11]曾杰,张华.基于最小二乘支持向量机的风速预测模型[J].电网技术,2009,33(18):144-147.Zeng Jie,Zhang Hua.A wind speed forecasting model based on leastsquares support vector machine[J].Power System Technology,2009,33(18):114-117(in Chinese).
- [12]邰能灵,侯志俭,李涛,等.基于小波分析的电力系统短期负荷预测方法[J].中国电机工程学报,2003,23(1):45-50.Tai Nengling,Hou Zhijian,Li Tao,et al.New principle based onwavelet transform for power system short-term loadforecasting[J].Proceedings of the CSEE,2003,23(1):45-50(inChinese).
- [13]Morra J H,Zhuowen T U,Apostolova L G.Comparison of adaboostand support vector machines for detecting Alzheimer's diseasethrough automated hippocampal segmentation[J].IEEE Trans onMedical Imaging,2010,29(1):30-43.
- [14]高琳,高峰,管晓宏,等.电力系统短期负荷预测的多神经网络Boosting集成模型[J].西安交通大学学报,2004,38(10):1026-1030.Gao Lin,Gao Feng,Guan Xiaohong,et al.Neural networks ensemblemodel based on boosting algorithm for short-term loadforecasting[J].Journal of Xi’an Jiaotong University,2004,38(10):1026-1030(in Chinese).
- [15]陈淑燕,王炜,瞿高峰,等.交通量预测的神经网络集成方法[J].公路交通科技,2004,21(12):80-83.Chen Shuyan,Wang Wei,Qu Gaofeng,et al.Traffic flow forecastbased on neural network ensemble[J].Journal of Highway andTransportation Research and Development,2004,21(12):80-83(inChinese).