多算法多模型与在线第二次学习结合的短期电力负荷预测方法
|
摘要
为了提高短期电力负荷预测精度,首次提出多算法多模型与在线第二次学习结合的预测方法。首先,利用互信息方法和统计方法对输入变量进行选择;然后,通过Bootstrap方法对数据集进行多样性采样,利用多个不同的人工智能算法和机器学习算法训练得到多个差异化较大的异构预测模型;最后,用每个待预测时刻最近一段时间的实际负荷值、第一次学习生成的多异构预测模型的负荷预测值构成新训练数据集,对新训练数据集进行在线第二次学习,得到最终预测结果。对中国广州市负荷进行预测研究,与最优单模型、单算法多模型和多算法单模型相比,在每日总负荷预测中,全年平均绝对百分误差(MAPE)分别下降了21.07%、7.64%和5.00%,在每日峰值负荷预测中,全年MAPE分别下降了16.02%、7.60%和13.14%。实验结果表明,推荐方法有效地提高了负荷预测精度,有利于智能电网实现节能降耗、调度精细化管理和电网安全预警。
In order to improve the forecasting accuracy of the short-term power load, a forecasting method combining multi-algorithm & multi-model and online second learning was newly proposed. First, the input variables were selected by using mutual information and statistical information and a dataset was constructed. Then, multiple training sets were generated by performing diverse sampling with bootstrap on the original training set. Multiple models were obtained using different artificial intelligence and machine-learning algorithms. Finally, the offline second-learning method was improved. A new training set was generated using the actual load, and the multi-model forecasts for recent period within the forecasted time,which is trained by online second learning to obtain the final forecasting results. The load in Guangzhou, China was studied.Compared to the optimal single-model, single-algorithm & multi-model and multi-algorithm & single-model, Mean Absolute Percentage Error( MAPE) of the proposed model was reduced by 21. 07%, 7. 64% and 5. 00%, respectively, in the daily total load forecasting, and by 16. 02%, 7. 60%, and 13. 14%, respectively, in the daily peak load forecasting. The experimental results show that the proposed method can improve the prediction accuracy of the power load, reduce costs,implement optimal scheduling management, and ensure security with early warnings in smart grids.
In order to improve the forecasting accuracy of the short-term power load, a forecasting method combining multi-algorithm & multi-model and online second learning was newly proposed. First, the input variables were selected by using mutual information and statistical information and a dataset was constructed. Then, multiple training sets were generated by performing diverse sampling with bootstrap on the original training set. Multiple models were obtained using different artificial intelligence and machine-learning algorithms. Finally, the offline second-learning method was improved. A new training set was generated using the actual load, and the multi-model forecasts for recent period within the forecasted time,which is trained by online second learning to obtain the final forecasting results. The load in Guangzhou, China was studied.Compared to the optimal single-model, single-algorithm & multi-model and multi-algorithm & single-model, Mean Absolute Percentage Error( MAPE) of the proposed model was reduced by 21. 07%, 7. 64% and 5. 00%, respectively, in the daily total load forecasting, and by 16. 02%, 7. 60%, and 13. 14%, respectively, in the daily peak load forecasting. The experimental results show that the proposed method can improve the prediction accuracy of the power load, reduce costs,implement optimal scheduling management, and ensure security with early warnings in smart grids.
引文
[1]DUDEK G.Pattern-based local linear regression models for shortterm load forecasting[J].Electric Power Systems Research,2016,130:139-147.
[2]CHO M Y,HWANG J C,CHEN C S.Customer short term load forecasting by using ARIMA transfer function model[C]//Proceedings of the 1995 International Conference on Energy Management and Power Delivery.Piscataway,NJ:IEEE,1995:317-322.
[3]LI S,WANG P,GOEL L.Short-term load forecasting by wavelet transform and evolutionary extreme learning machine[J].Electric Power Systems Research,2015,122:96-103.
[4]XIAO Z,YE S J,ZHONG B,et al.BP neural network with rough set for short term load forecasting[J].Expert Systems with Applications,2009,36(1):273-279.
[5]师彪,李郁侠,于新花,等.改进粒子群——BP神经网络模型的短期电力负荷预测[J].计算机应用,2009,29(4):1036-1039.(SHI B,LI Y X,YU X H,et al.Short-term load forecast based on modified particle swarm optimizer and back propagation neural network model[J].Journal of Computer Applications,2009,29(4):1036-1039.)
[6]郎坤,张明媛,袁永博.基于迭代误差补偿的核极端学习机模型在短期电力负荷预测中的应用[J].计算机应用,2015,35(7):2083-2087.(LANG K,ZHANG M Y,YUAN Y B.Application of extreme learning machine with kernels model based on iterative error correction in short term electricity load forecasting[J].Journal of Computer Applications,2015,35(7):2083-2087.)
[7]吴倩红,高军,侯广松,等.实现影响因素多源异构融合的短期负荷预测支持向量机算法[J].电力系统自动化,2016,40(15):67-72.(WU Q H,GAO J,HOU G S,et al.Short-term load forecasting support vector machine algorithm based on multi-source heterogeneous fusion of load factors[J].Automation of Electric Power Systems,2016,40(15):67-72.)
[8]CHEN Y H,HONG W C,SHEN W,et al.Electric load forecasting based on a least squares support vector machine with fuzzy time series and global harmony search algorithm[J].Energies,2016,9(2):7-82.
[9]LIN W M,TU C S,YANG R F,et al.Particle swarm optimisation aided least-square support vector machine for load forecast with spikes[J].IET Generation,Transmission&Distribution,2016,10(5):1145-1153.
[10]胡时雨,罗滇生,阳霜,等.基于多变量LS-SVM和模糊循环推理系统的负荷预测[J].计算机应用,2015,35(2):595-600.(HU S Y,LUO D S,YANG S,et al.Load forecasting based on multi-variable LS-SVM and fuzzy recursive inference system[J].Journal of Computer Applications,2015,35(2):595-600.)
[11]PAPADOPOULOS S,KARAKATSANIS I.Short-term electricity load forecasting using time series and ensemble learning methods[C]//Proceedings of the 2015 IEEE Power and Energy Conference at Illinois.Piscataway:IEEE,2015:1-6.
[12]WU X Y,HE J H,YIP T,et al.A two-stage random forest method for short-term load forecasting[C]//Proceedings of the 2016IEEE Power and Energy Society General Meeting.Piscataway,NJ:IEEE,2016:1-5.
[13]TAIEB S B,HYNDMAN R J.A gradient boosting approach to the Kaggle load forecasting competition[J].International Journal of Forecasting,2014,30(2):382-394.
[14]MOHAN A,CHEN Z,WEINBERGER K.Web-search ranking with initialized gradient boosted regression trees[C]//Proceedings of the 2010 International Conference on Yahoo!Learning to Rank Challenge.Haifa,Israel:JMLR,2010,14:77-89.
[15]LI N,JIANG Y,ZHOU Z H.Multi-label selective ensemble[C]//Proceedings of the 12th International Workshop on Multiple Classifier Systems.Berlin:Springer,2015:76-88.
[16]ZHOU Z H,WU J X,TANG W.Ensembling neural networks:many could be better than all[J].Artificial Intelligence,2002,137(1):239-263.
[17]WANG B,CHIANG H D.ELITE:ensemble of optimal inputpruned neural networks using TRUST-TECH[J].IEEE Transactions on Neural Networks,2011,22(1):96-109.
[18]LI S,GOEL L,WANG P.An ensemble approach for short-term load forecasting by extreme learning machine[J].Applied Energy,2016,170:22-29.
[19]CHEN L G,CHIANG H D,DONG N,et al.Group-based chaos genetic algorithm and non-linear ensemble of neural networks for short-term load forecasting[J].IET Generation Transmission&Distribution,2016,10(6):1440-1447.
[20]RIBEIRO G T,GRITTI M C,AYALA H V H,et al.Short-term load forecasting using wavenet ensemble approaches[C]//Proceedings of the 2016 International Joint Conference on Neural Networks.Piscataway,NJ:IEEE,2016:727-734.
[21]SIWEK K,OSOWSKI S,SZUPILUK R.Ensemble neural network approach for accurate load forecasting in a power system[J].International Journal of Applied Mathematics&Computer Science,2009,19(2):303-315.
[22]GRMANOVA G,LAURINEC P,ROZINAJOVA V,et al.Incremental ensemble learning for electricity load forecasting[J].Acta Polytechnica Hungarica,2016,13(2):97-117.
[23]DUDEK G.Heterogeneous ensembles for short-term electricity demand forecasting[C]//Proceedings of the 2016 17th International Scientific Conference on Electric Power Engineering.Piscataway,NJ:IEEE,2016:1-6.
[24]SUYKENS J A K,VANDEWALLE J.Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[25]BREIMAN L.Random forests[J].Machine Learning,2001,45(1):5-32.
[26]FRIEDMAN J H.Greedy function approximation:a gradient boosting machine[J].The Annals of Statistics,2001,29(5):1189-1232.
[27]HECHTNIELSEN R.Theory of the back propagation neural network[C]//Proceedings of the 1989 International Joint Conference on Neural Networks.Piscataway,NJ:IEEE,1989:593-605.
[28]GUILLEN A,HERRERA L J,RUBIO G,et al.New method for instance or prototype selection using mutual information in time series prediction[J].Neurocomputing,2010,73(10):2030-2038.
[29]BONNLANDER B V,WEIGEND A S.Selecting input variables using mutual information and nonparametric density estimation[EB/OL].[2016-11-20].http://pdfs.semanticscholar.org/0cbb/68b53c3a079545790e1e97e9f14bb4d613a9.pdf.
[2]CHO M Y,HWANG J C,CHEN C S.Customer short term load forecasting by using ARIMA transfer function model[C]//Proceedings of the 1995 International Conference on Energy Management and Power Delivery.Piscataway,NJ:IEEE,1995:317-322.
[3]LI S,WANG P,GOEL L.Short-term load forecasting by wavelet transform and evolutionary extreme learning machine[J].Electric Power Systems Research,2015,122:96-103.
[4]XIAO Z,YE S J,ZHONG B,et al.BP neural network with rough set for short term load forecasting[J].Expert Systems with Applications,2009,36(1):273-279.
[5]师彪,李郁侠,于新花,等.改进粒子群——BP神经网络模型的短期电力负荷预测[J].计算机应用,2009,29(4):1036-1039.(SHI B,LI Y X,YU X H,et al.Short-term load forecast based on modified particle swarm optimizer and back propagation neural network model[J].Journal of Computer Applications,2009,29(4):1036-1039.)
[6]郎坤,张明媛,袁永博.基于迭代误差补偿的核极端学习机模型在短期电力负荷预测中的应用[J].计算机应用,2015,35(7):2083-2087.(LANG K,ZHANG M Y,YUAN Y B.Application of extreme learning machine with kernels model based on iterative error correction in short term electricity load forecasting[J].Journal of Computer Applications,2015,35(7):2083-2087.)
[7]吴倩红,高军,侯广松,等.实现影响因素多源异构融合的短期负荷预测支持向量机算法[J].电力系统自动化,2016,40(15):67-72.(WU Q H,GAO J,HOU G S,et al.Short-term load forecasting support vector machine algorithm based on multi-source heterogeneous fusion of load factors[J].Automation of Electric Power Systems,2016,40(15):67-72.)
[8]CHEN Y H,HONG W C,SHEN W,et al.Electric load forecasting based on a least squares support vector machine with fuzzy time series and global harmony search algorithm[J].Energies,2016,9(2):7-82.
[9]LIN W M,TU C S,YANG R F,et al.Particle swarm optimisation aided least-square support vector machine for load forecast with spikes[J].IET Generation,Transmission&Distribution,2016,10(5):1145-1153.
[10]胡时雨,罗滇生,阳霜,等.基于多变量LS-SVM和模糊循环推理系统的负荷预测[J].计算机应用,2015,35(2):595-600.(HU S Y,LUO D S,YANG S,et al.Load forecasting based on multi-variable LS-SVM and fuzzy recursive inference system[J].Journal of Computer Applications,2015,35(2):595-600.)
[11]PAPADOPOULOS S,KARAKATSANIS I.Short-term electricity load forecasting using time series and ensemble learning methods[C]//Proceedings of the 2015 IEEE Power and Energy Conference at Illinois.Piscataway:IEEE,2015:1-6.
[12]WU X Y,HE J H,YIP T,et al.A two-stage random forest method for short-term load forecasting[C]//Proceedings of the 2016IEEE Power and Energy Society General Meeting.Piscataway,NJ:IEEE,2016:1-5.
[13]TAIEB S B,HYNDMAN R J.A gradient boosting approach to the Kaggle load forecasting competition[J].International Journal of Forecasting,2014,30(2):382-394.
[14]MOHAN A,CHEN Z,WEINBERGER K.Web-search ranking with initialized gradient boosted regression trees[C]//Proceedings of the 2010 International Conference on Yahoo!Learning to Rank Challenge.Haifa,Israel:JMLR,2010,14:77-89.
[15]LI N,JIANG Y,ZHOU Z H.Multi-label selective ensemble[C]//Proceedings of the 12th International Workshop on Multiple Classifier Systems.Berlin:Springer,2015:76-88.
[16]ZHOU Z H,WU J X,TANG W.Ensembling neural networks:many could be better than all[J].Artificial Intelligence,2002,137(1):239-263.
[17]WANG B,CHIANG H D.ELITE:ensemble of optimal inputpruned neural networks using TRUST-TECH[J].IEEE Transactions on Neural Networks,2011,22(1):96-109.
[18]LI S,GOEL L,WANG P.An ensemble approach for short-term load forecasting by extreme learning machine[J].Applied Energy,2016,170:22-29.
[19]CHEN L G,CHIANG H D,DONG N,et al.Group-based chaos genetic algorithm and non-linear ensemble of neural networks for short-term load forecasting[J].IET Generation Transmission&Distribution,2016,10(6):1440-1447.
[20]RIBEIRO G T,GRITTI M C,AYALA H V H,et al.Short-term load forecasting using wavenet ensemble approaches[C]//Proceedings of the 2016 International Joint Conference on Neural Networks.Piscataway,NJ:IEEE,2016:727-734.
[21]SIWEK K,OSOWSKI S,SZUPILUK R.Ensemble neural network approach for accurate load forecasting in a power system[J].International Journal of Applied Mathematics&Computer Science,2009,19(2):303-315.
[22]GRMANOVA G,LAURINEC P,ROZINAJOVA V,et al.Incremental ensemble learning for electricity load forecasting[J].Acta Polytechnica Hungarica,2016,13(2):97-117.
[23]DUDEK G.Heterogeneous ensembles for short-term electricity demand forecasting[C]//Proceedings of the 2016 17th International Scientific Conference on Electric Power Engineering.Piscataway,NJ:IEEE,2016:1-6.
[24]SUYKENS J A K,VANDEWALLE J.Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[25]BREIMAN L.Random forests[J].Machine Learning,2001,45(1):5-32.
[26]FRIEDMAN J H.Greedy function approximation:a gradient boosting machine[J].The Annals of Statistics,2001,29(5):1189-1232.
[27]HECHTNIELSEN R.Theory of the back propagation neural network[C]//Proceedings of the 1989 International Joint Conference on Neural Networks.Piscataway,NJ:IEEE,1989:593-605.
[28]GUILLEN A,HERRERA L J,RUBIO G,et al.New method for instance or prototype selection using mutual information in time series prediction[J].Neurocomputing,2010,73(10):2030-2038.
[29]BONNLANDER B V,WEIGEND A S.Selecting input variables using mutual information and nonparametric density estimation[EB/OL].[2016-11-20].http://pdfs.semanticscholar.org/0cbb/68b53c3a079545790e1e97e9f14bb4d613a9.pdf.