基于相似日和人工神经网络的风电功率短期预测研究
|
摘要
随着地球上的化石燃料的不断消耗,风能作为清洁、安全的能源正在改变着全球能源结构,由于自然风具有随机性、波动性和不可控制性,造成风电场在发电时,发电功率产生巨大的波动,为了提高风电功率预测精度,采用人工神经网络和相似日的方法,以云南某电场风电场发电功率的数据为例,建立模型对风电功率进行了短期预测,研究得出:该方法能够有效地对风电场功率进行预测;与传统BP神经网络相比而言,基于人工神经网络和相似日的方法具有很强的非线性学习的能力,对提高高精度风电场输出功率的预测很有帮助;基于人工神经网络和相似日的方法预测误差概率,误差概率分布符合正态分布,可以作为风电场发电功率误差的置信区间估计和预测的依据,研究为风电功率的预测提供了一定的借鉴意义。
With the continuous consumption of fossil fuels on the earth,wind energy was changing the global energy structure as a clean and safe energy source. Due to the randomness,volatility and uncontrollability of natural winds,the power generation of wind farms generates huge power. In order to improve the prediction accuracy of wind power,the artificial neural network and similar day method were used to take the data of wind power generation of an electric field wind farm in Yunnan as an example. The model was used to predict the wind power in a short time. The research results showed that the method can effectively predicting the power of wind farms; compared with traditional BP neural networks,artificial neural networks and similar-day methods have strong nonlinear learning ability,and were very predictive of improving the output power of high-precision wind farms. it based on artificial neural network and similar day method to predict the error probability,the error probability distribution was in accordance with the normal distribution,which could be used as the basis for the confidence interval estimation and prediction of wind farm power generation error. Research provides a reference for the prediction of wind power.
With the continuous consumption of fossil fuels on the earth,wind energy was changing the global energy structure as a clean and safe energy source. Due to the randomness,volatility and uncontrollability of natural winds,the power generation of wind farms generates huge power. In order to improve the prediction accuracy of wind power,the artificial neural network and similar day method were used to take the data of wind power generation of an electric field wind farm in Yunnan as an example. The model was used to predict the wind power in a short time. The research results showed that the method can effectively predicting the power of wind farms; compared with traditional BP neural networks,artificial neural networks and similar-day methods have strong nonlinear learning ability,and were very predictive of improving the output power of high-precision wind farms. it based on artificial neural network and similar day method to predict the error probability,the error probability distribution was in accordance with the normal distribution,which could be used as the basis for the confidence interval estimation and prediction of wind farm power generation error. Research provides a reference for the prediction of wind power.
引文
[1]孟洋洋,卢继平,孙华利,等.基于相似日和人工神经网络的风电功率短期预测[J].电网技术,2010,34(12):163-167.Meng Yangyang,Lu Jiping,SunHuali,et al. Short-term prediction of wind power based on similar day and artificial neural network[J]. Power System Technology,2010,34(12):163-167.
[2]彭怀午,刘方锐,杨晓峰.基于人工神经网络的风电功率短期预测研究[J].华东电力,2009,37(11):130-133.Peng Huaiwu,Liu Fangrui,Yang Xiaofeng. Research on short-term wind power forecasting based on artificial neural network[J]. East China Electric Power,2009,37(11):130-133.
[3]杨永强,尹成全,马腾飞,等.云南风力发电特性探析[J].云南电力技术,2016,44(2):10-13.Yang Yongqiang,Yin Chengquan,Ma Tengfei,et al. Analysis on the characteristics of wind power generation in Yunnan[J]. Yunnan Electric Power,2016,44(2):10-13.
[4]范高锋,王伟胜,刘纯.基于人工神经网络的风电功率短期预测系统[J].电网技术,2008,32(22):72-76.Fan Gaofeng,Wang Weisheng,Liu Chun. Wind power forecasting system based on artificial neural network[J]. Power System Technology,2008,32(22):72-76.
[5]黄辰,吴峻青.基于人工神经网络的短期风电功率预测[J].华东电力,2014,42(7):1408-1410.Huang Chen,Wu Junqing. Short-term wind power prediction based on artificial neural network[J]. East China Electric Power,2014,42(7):1408-1410.
[6]范高锋,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 power forecast based on artificial neural network[J]. Proceedings of the CSEE,2008,28(34):118-123.
[7]张宜阳,严欢.基于分段分层相似日搜索和自适应脊波神经网络的风电功率多步预测[J].电网与清洁能源,2015,31(4):124-131.Zhang Yiyang,Yan Huan. Wind power multi-step prediction based on segmented layered similarity day search and adaptive ridgelet neural network[J]. Power System and Clean Energy,2015,31(4):124-131.
[8]丁明,王磊,毕锐.基于改进BP神经网络的光伏发电系统输出功率短期预测模型[J].电力系统保护与控制,2012,40(11):93-99.Ding Ming,Wang Lei,Bi Rui. Short-term prediction model of photovoltaic power generation system output power based on improved BP neural network[J]. Power System Protection and Control,2012,40(11):93-99.
[9]白永祥,房大中,侯佑华,等.内蒙古电网区域风电功率预测系统[J].电网技术,2010(10):157-162.Bai Yongxiang,Fang Dazhong,Hou Youhua,et al. Inner Mongolia power grid regional wind power forecasting system[J]. Power System Technology,2010(10):157-162.
[10]周松林,茆美琴,苏建徽.基于主成分分析与人工神经网络的风电功率预测[J].电网技术,2011,35(9):128-132.Zhou Songlin,Mou Meiqin,Su Jianhui. Wind power forecast based on principal component analysis and artificial neural network[J].Power System Technology,2011,35(9):128-132.
[11]姜勇,卢毅.基于相似日的神经网络短期负荷预测方法[J].电力系统及其自动化学报,2001,23(6):16-18.Jiang Yong,Lu Yi. Neural network short-term load forecasting method based on similar days[J]. Proceedings of the CSU-EPSA,2001,23(6):16-18.
[12]董存,高爽,郝颖,等.基于聚类分析的风电场短期功率预测方法研究[J]. Sustainable Energy,2015,5(3):17-23.Dong Cun,Gao Shuang,Hao Ying,et al. Research on wind farm short-term power forecasting method based on clustering analysis[J]. Sustainable Energy,2015,5(3):17-23.
[13]牛东晓,魏亚楠.基于FHNN相似日聚类自适应权重的短期电力负荷组合预测[J].电力系统自动化,2013,37(3):54-57.Niu Dongxiao,Wei Yanan. Short-term combined load forecasting based on adaptive weights of FHNN similar-day clustering[J]. Automation of Electric Power Systems,2013,37(3):54-57.
[14]周晖,王玮,秦海超,等.基于多时段气象数据判断相似日的日负荷曲线预测研究[J].电力系统保护与控制,2005,33(23):41-45.Zhou Hui,Wang Wei,Qin Haichao,et al. Research on daily load curve prediction based on multi-period meteorological data to determine similar day[J]. Power System Protection and Control,2005,33(23):41-45.
[15]李振然,贾旭彩.基于人工神经网络的短期负荷预测[J].广西电力,2002,25(4):7-10.Li Zhenran,Jia Xucai. Short-term load forecasting based on artificial neural network[J]. Guangxi Electric Power,2002,25(4):7-10.
[16]陈素华,贺凌霄.电网短期负荷预测管理系统的应用研究[J].电气传动,2013(s1):186-189.Chen Suhua,He Lingxiao. Application of power grid short-term load forecasting management system[J]. Electric Drive,2013(s1):186-189.
[17]董秀成,李芹,许强. BP算法在电力系统短期负荷预测中的应用[J].仪器仪表学报,2003,24(S2):379-381.Dong Xiucheng,Li Qin,Xu Qiang. Application of BP algorithm in power system short-term load forecasting[J]. Chinese Journal of Scientific Instrument,2003,24(S2):379-381.
[18]嵇灵,牛东晓,汪鹏.基于相似日聚类和贝叶斯神经网络的光伏发电功率预测研究[J].中国管理科学,2015,23(3):118-122.Ji Ling,Niu Dongxiao,Wang Peng. Research on photovoltaic power forecast based on similar day clustering and bayesian neural network[J]. Chinese Journal of Management Science,2015,23(3):118-122.
[19]倪鹏,孙富荣.基于相似日和风速连续性的风电场短期功率预测[J].山东电力技术,2016,43(11):39-43.Ni Peng,Sun Furong. Wind farm short-term power forecast based on similarity day and wind speed continuity[J]. Shandong Electric Power,2016,43(11):39-43.
[20]张佳宁.基于神经网络及相似日的城市交通流量预测模型研究[J].工业控制计算机,2016,29(4):71-72.Zhang Jianing. Urban traffic flow prediction model based on neural network and similar day[J]. Industrial Control Computer,2016,29(4):71-72.
[21]姜勇.电力系统短期负荷预测的模糊神经网络方法[J].电力系统保护与控制,2002,30(7):1-3.Jiang Yong. Short-term load forecasting of power system based on fuzzy neural network[J]. Power System Protection and Control,2002,30(7):1-3.
[22]陈通,孙国强,卫志农,等.基于Spiking神经网络的光伏系统发电功率预测[J].电力系统及其自动化学报,2017,29(6):7-12.Chen Tong,Sun Guoqiang,Wei Zhinong,et al. PV power generation power prediction based on Spiking neural network[J]. Proceedings of the CSU-EPSA,2017,29(6):7-12.
[23]马立新,尹晶晶,郑晓栋.基于特征提取相似日的ELM短期负荷预测研究[J].电子科技,2015,28(12):22-25.Ma Lixin,Yin Jingjing,Zheng Xiaodong. Research on short-term load forecasting of ELM based on feature extraction similarity day[J]. Electronic Science and Technology,2015,28(12):22-25.
[24]张颖,张瑞,高中文.基于相似日的神经网络短期负荷预测方法[J].黑龙江大学自然科学学报,2003,20(2):59-61.Zhang Ying,Zhang Rui,Gao Zhongwen. Neural network short-term load forecasting method based on similar days[J]. Journal of Natural Science of Heilongjiang University,2003,20(2):59-61.
[25]宋人杰,刘福盛,马冬梅,等.基于相似日和WNN的光伏发电功率超短期预测模型[J].电测与仪表,2017,54(7):75-80.Song Renjie,Liu Fusheng,Ma Dongmei,et al. Photovoltaic power ultrashort-term forecasting model based on similar day and WNN[J]. Electrical Measurement&Instrumentation,2017,54(7):75-80.
[26]杜欣慧,张岭,毕艳华.采用自适应神经网络进行短期负荷预测[J].电子测量与仪器学报,2004,18(S2):632-635.Du Xinhui,Zhang Ling,Bi Yanhua. Adaptive neural network for short-term load forecasting[J]. Journal of Electronic Measurement and Instrumentation,2004,18(S2):632-635.
[27]吴潇雨,和敬涵,张沛,等.基于灰色投影改进随机森林算法的电力系统短期负荷预测[J].电力系统自动化,2015,39(12):50-55.Wu Xiaoyu,He Jinghan,Zhang Pei,et al. Short term load forecasting of power system based on grey projection improved random forest algorithm[J]. Automation of Electric Power Systems,2015,39(12):50-55.
[28]刘沛汉,袁铁江,梅生伟,等.基于遗传算法优化神经网络的光伏电站短期功率预测[J].水电能源科学,2016(1):211-214.Liu Peiquan,Yuan Tiejiang,Mei Shengwei,et al. Short-term power prediction of photovoltaic power station based on genetic algorithm and optimization of neural network[J]. Water Resources and Power,2016(1):211-214.
[29]罗小兰,刘士荣,郑凌蔚,等.基于相似日的光伏发电短期预报模型[J].杭州电子科技大学学报,2015,35(5):74-78.Luo Xiaolan,Liu Shirong,Zheng Lingwei,et al. Photovoltaic power generation short-term forecasting model based on similar days[J].Journal of Hangzhou Dianzi University(Natural Sciences),2015,35(5):74-78.
[30]王晓兰,葛鹏江.基于相似日和径向基函数神经网络的光伏阵列输出功率预测[J].电力自动化设备,2013,33(1):100-103.Wang Xiaolan,Ge Pengjiang. PV array output power prediction based on similar day and radial basis function neural network[J].Electric Power Automation Equipment,2013,33(1):100-103.
[31]李鹏梅,臧传治,王侃侃.基于相似日和神经网络的光伏发电预测[J].可再生能源,2013,31(10):1-4.Li Pengmei,Zang Chuanzhi,Wang Kankan. PV generation forecast based on similar days and neural networks[J]. Renewable Energy Resources,2013,31(10):1-4.
[32]罗勇,郑金,宁美凤.基于相似日搜索的PSO-WNN组合模型在短期电力负荷预测中的应用[J].信息与控制,2013,42(3):371-376.Luo Yong,Zheng Jin,Ning Meifeng. Application of PSO-WNN combined model based on similar day search in short-term load forecasting[J]. Information and Control,2013,42(3):371-376.
[33]钟春霞.基于相似日选择算法和Elman神经网络的光伏输出功率预测[J].南京工程学院学报(自然科学版),2016,14(1):42-47.Zhong Chunxia. PV output power prediction based on similar-day selection algorithm and Elman neural network[J]. Journal of Nanjing Institute of Technology(Natural Science Edition),2016,14(1):42-47.
[2]彭怀午,刘方锐,杨晓峰.基于人工神经网络的风电功率短期预测研究[J].华东电力,2009,37(11):130-133.Peng Huaiwu,Liu Fangrui,Yang Xiaofeng. Research on short-term wind power forecasting based on artificial neural network[J]. East China Electric Power,2009,37(11):130-133.
[3]杨永强,尹成全,马腾飞,等.云南风力发电特性探析[J].云南电力技术,2016,44(2):10-13.Yang Yongqiang,Yin Chengquan,Ma Tengfei,et al. Analysis on the characteristics of wind power generation in Yunnan[J]. Yunnan Electric Power,2016,44(2):10-13.
[4]范高锋,王伟胜,刘纯.基于人工神经网络的风电功率短期预测系统[J].电网技术,2008,32(22):72-76.Fan Gaofeng,Wang Weisheng,Liu Chun. Wind power forecasting system based on artificial neural network[J]. Power System Technology,2008,32(22):72-76.
[5]黄辰,吴峻青.基于人工神经网络的短期风电功率预测[J].华东电力,2014,42(7):1408-1410.Huang Chen,Wu Junqing. Short-term wind power prediction based on artificial neural network[J]. East China Electric Power,2014,42(7):1408-1410.
[6]范高锋,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 power forecast based on artificial neural network[J]. Proceedings of the CSEE,2008,28(34):118-123.
[7]张宜阳,严欢.基于分段分层相似日搜索和自适应脊波神经网络的风电功率多步预测[J].电网与清洁能源,2015,31(4):124-131.Zhang Yiyang,Yan Huan. Wind power multi-step prediction based on segmented layered similarity day search and adaptive ridgelet neural network[J]. Power System and Clean Energy,2015,31(4):124-131.
[8]丁明,王磊,毕锐.基于改进BP神经网络的光伏发电系统输出功率短期预测模型[J].电力系统保护与控制,2012,40(11):93-99.Ding Ming,Wang Lei,Bi Rui. Short-term prediction model of photovoltaic power generation system output power based on improved BP neural network[J]. Power System Protection and Control,2012,40(11):93-99.
[9]白永祥,房大中,侯佑华,等.内蒙古电网区域风电功率预测系统[J].电网技术,2010(10):157-162.Bai Yongxiang,Fang Dazhong,Hou Youhua,et al. Inner Mongolia power grid regional wind power forecasting system[J]. Power System Technology,2010(10):157-162.
[10]周松林,茆美琴,苏建徽.基于主成分分析与人工神经网络的风电功率预测[J].电网技术,2011,35(9):128-132.Zhou Songlin,Mou Meiqin,Su Jianhui. Wind power forecast based on principal component analysis and artificial neural network[J].Power System Technology,2011,35(9):128-132.
[11]姜勇,卢毅.基于相似日的神经网络短期负荷预测方法[J].电力系统及其自动化学报,2001,23(6):16-18.Jiang Yong,Lu Yi. Neural network short-term load forecasting method based on similar days[J]. Proceedings of the CSU-EPSA,2001,23(6):16-18.
[12]董存,高爽,郝颖,等.基于聚类分析的风电场短期功率预测方法研究[J]. Sustainable Energy,2015,5(3):17-23.Dong Cun,Gao Shuang,Hao Ying,et al. Research on wind farm short-term power forecasting method based on clustering analysis[J]. Sustainable Energy,2015,5(3):17-23.
[13]牛东晓,魏亚楠.基于FHNN相似日聚类自适应权重的短期电力负荷组合预测[J].电力系统自动化,2013,37(3):54-57.Niu Dongxiao,Wei Yanan. Short-term combined load forecasting based on adaptive weights of FHNN similar-day clustering[J]. Automation of Electric Power Systems,2013,37(3):54-57.
[14]周晖,王玮,秦海超,等.基于多时段气象数据判断相似日的日负荷曲线预测研究[J].电力系统保护与控制,2005,33(23):41-45.Zhou Hui,Wang Wei,Qin Haichao,et al. Research on daily load curve prediction based on multi-period meteorological data to determine similar day[J]. Power System Protection and Control,2005,33(23):41-45.
[15]李振然,贾旭彩.基于人工神经网络的短期负荷预测[J].广西电力,2002,25(4):7-10.Li Zhenran,Jia Xucai. Short-term load forecasting based on artificial neural network[J]. Guangxi Electric Power,2002,25(4):7-10.
[16]陈素华,贺凌霄.电网短期负荷预测管理系统的应用研究[J].电气传动,2013(s1):186-189.Chen Suhua,He Lingxiao. Application of power grid short-term load forecasting management system[J]. Electric Drive,2013(s1):186-189.
[17]董秀成,李芹,许强. BP算法在电力系统短期负荷预测中的应用[J].仪器仪表学报,2003,24(S2):379-381.Dong Xiucheng,Li Qin,Xu Qiang. Application of BP algorithm in power system short-term load forecasting[J]. Chinese Journal of Scientific Instrument,2003,24(S2):379-381.
[18]嵇灵,牛东晓,汪鹏.基于相似日聚类和贝叶斯神经网络的光伏发电功率预测研究[J].中国管理科学,2015,23(3):118-122.Ji Ling,Niu Dongxiao,Wang Peng. Research on photovoltaic power forecast based on similar day clustering and bayesian neural network[J]. Chinese Journal of Management Science,2015,23(3):118-122.
[19]倪鹏,孙富荣.基于相似日和风速连续性的风电场短期功率预测[J].山东电力技术,2016,43(11):39-43.Ni Peng,Sun Furong. Wind farm short-term power forecast based on similarity day and wind speed continuity[J]. Shandong Electric Power,2016,43(11):39-43.
[20]张佳宁.基于神经网络及相似日的城市交通流量预测模型研究[J].工业控制计算机,2016,29(4):71-72.Zhang Jianing. Urban traffic flow prediction model based on neural network and similar day[J]. Industrial Control Computer,2016,29(4):71-72.
[21]姜勇.电力系统短期负荷预测的模糊神经网络方法[J].电力系统保护与控制,2002,30(7):1-3.Jiang Yong. Short-term load forecasting of power system based on fuzzy neural network[J]. Power System Protection and Control,2002,30(7):1-3.
[22]陈通,孙国强,卫志农,等.基于Spiking神经网络的光伏系统发电功率预测[J].电力系统及其自动化学报,2017,29(6):7-12.Chen Tong,Sun Guoqiang,Wei Zhinong,et al. PV power generation power prediction based on Spiking neural network[J]. Proceedings of the CSU-EPSA,2017,29(6):7-12.
[23]马立新,尹晶晶,郑晓栋.基于特征提取相似日的ELM短期负荷预测研究[J].电子科技,2015,28(12):22-25.Ma Lixin,Yin Jingjing,Zheng Xiaodong. Research on short-term load forecasting of ELM based on feature extraction similarity day[J]. Electronic Science and Technology,2015,28(12):22-25.
[24]张颖,张瑞,高中文.基于相似日的神经网络短期负荷预测方法[J].黑龙江大学自然科学学报,2003,20(2):59-61.Zhang Ying,Zhang Rui,Gao Zhongwen. Neural network short-term load forecasting method based on similar days[J]. Journal of Natural Science of Heilongjiang University,2003,20(2):59-61.
[25]宋人杰,刘福盛,马冬梅,等.基于相似日和WNN的光伏发电功率超短期预测模型[J].电测与仪表,2017,54(7):75-80.Song Renjie,Liu Fusheng,Ma Dongmei,et al. Photovoltaic power ultrashort-term forecasting model based on similar day and WNN[J]. Electrical Measurement&Instrumentation,2017,54(7):75-80.
[26]杜欣慧,张岭,毕艳华.采用自适应神经网络进行短期负荷预测[J].电子测量与仪器学报,2004,18(S2):632-635.Du Xinhui,Zhang Ling,Bi Yanhua. Adaptive neural network for short-term load forecasting[J]. Journal of Electronic Measurement and Instrumentation,2004,18(S2):632-635.
[27]吴潇雨,和敬涵,张沛,等.基于灰色投影改进随机森林算法的电力系统短期负荷预测[J].电力系统自动化,2015,39(12):50-55.Wu Xiaoyu,He Jinghan,Zhang Pei,et al. Short term load forecasting of power system based on grey projection improved random forest algorithm[J]. Automation of Electric Power Systems,2015,39(12):50-55.
[28]刘沛汉,袁铁江,梅生伟,等.基于遗传算法优化神经网络的光伏电站短期功率预测[J].水电能源科学,2016(1):211-214.Liu Peiquan,Yuan Tiejiang,Mei Shengwei,et al. Short-term power prediction of photovoltaic power station based on genetic algorithm and optimization of neural network[J]. Water Resources and Power,2016(1):211-214.
[29]罗小兰,刘士荣,郑凌蔚,等.基于相似日的光伏发电短期预报模型[J].杭州电子科技大学学报,2015,35(5):74-78.Luo Xiaolan,Liu Shirong,Zheng Lingwei,et al. Photovoltaic power generation short-term forecasting model based on similar days[J].Journal of Hangzhou Dianzi University(Natural Sciences),2015,35(5):74-78.
[30]王晓兰,葛鹏江.基于相似日和径向基函数神经网络的光伏阵列输出功率预测[J].电力自动化设备,2013,33(1):100-103.Wang Xiaolan,Ge Pengjiang. PV array output power prediction based on similar day and radial basis function neural network[J].Electric Power Automation Equipment,2013,33(1):100-103.
[31]李鹏梅,臧传治,王侃侃.基于相似日和神经网络的光伏发电预测[J].可再生能源,2013,31(10):1-4.Li Pengmei,Zang Chuanzhi,Wang Kankan. PV generation forecast based on similar days and neural networks[J]. Renewable Energy Resources,2013,31(10):1-4.
[32]罗勇,郑金,宁美凤.基于相似日搜索的PSO-WNN组合模型在短期电力负荷预测中的应用[J].信息与控制,2013,42(3):371-376.Luo Yong,Zheng Jin,Ning Meifeng. Application of PSO-WNN combined model based on similar day search in short-term load forecasting[J]. Information and Control,2013,42(3):371-376.
[33]钟春霞.基于相似日选择算法和Elman神经网络的光伏输出功率预测[J].南京工程学院学报(自然科学版),2016,14(1):42-47.Zhong Chunxia. PV output power prediction based on similar-day selection algorithm and Elman neural network[J]. Journal of Nanjing Institute of Technology(Natural Science Edition),2016,14(1):42-47.