邹政达,孙雅明,张智晟

• 1:天津大学电气与自动化工程学院
• 2:天津大学电气与自动化工程学院
• 3:天津大学电气与自动化工程学院 天津市南开区300072
• 4:天津市南开区300072

摘要(Abstract)：

为了克服BP算法收敛速度慢和易于陷入局部最小的不足,作者提出将蚁群优化算法用于短期负荷预测的递归神经网络模型学习算法,对实际负荷系统日、周预测的仿真测试表明,该模型能有效地提高短期负荷预测的精度,对工作日和休息日都具有良好的稳定性和适应能力,其预测性能明显优于基于BP算法的递归神经网络(BP-RNN)和基于遗传算法的递归神经网络(GA-RNN)。

关键词(KeyWords)： 蚁群优化算法;BP算法;递归神经网络;短期负荷预测;电力系统

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 邹政达,孙雅明,张智晟

Email:

DOI:

参考文献(References)：

• [1]DorigoM,ManiezzoV,ColorniA.Ant system: optimization by a colony of cooperating agents[J].IEEE Trans onSystems,Man andCybernetics,PartB,1996,26(1):29-41.
• [2]ColorniA,DorigoM,ManiezzoV.An investigation of some properties of an ant algorithm[C].Proceedings of theParallelProblemSolving fromNatureConference(PPSN?2)[A],Brussels,Belgium:ElsevierPublishing,1992:509-520.
• [3]DorigoM,CaroG.D.Ant colony optimization: a new meta-heuristic[C].Proceedings of the1999 congress onEvolutionary computation[A],Washington,DC:USA,1470-1477.
• [4]DorigoM,GambardellaL M.Ant colony system: a cooperative learning approach to the traveling salesman problem[J].IEEE Trans on,EvolutionaryComputation,1997,1(1):53-66.
• [5]El-KeibA A,SisworahardjoN S.Unit commitment using the ant colony search algorithm[C].PowerEngineering2002,Conference on,LargeEngineeringSystems[A].LESCOPE 02,Pages:2-6.
• [6]郝晋,石立宝,周家启,等.基于蚁群优化算法的机组最优投入[J].电网技术,2002,26(11):26-31.HaoJin,ShiLibao,ZhuoJiaqi et al.Optimal unit commitment based onAnt optimization algorithm[J].PowerSystemTechnology,2002,26(11):26-31.
• [7]ManiezzoV,ColorniA,DorigoM.The ant system applied to the quadralic assignment problem[J].Knowledge andDataEngineering,IEEE Transactions on,1999,11(5):769-778.
• [8]MerpP,FreislebenB.A comparison of memetic algorithms,tabu search,and ant colonies for the quadratic assignment problem[C].EvolutionaryComputation,CEC 99,Proceedings of the1999Congress on,Vol.3,Pages:2070.
• [9]陈根军,王磊,唐国庆,等.基于蚁群是优的配电网规划方法[J].电网技术,2003,27(3):71-75.ChenGenjun,WangLei,TangGuoqing et al.An ant colony optimization based method for power distribution network planning[J].PowerSystemTechnology,2003,27(3):71-75.
• [10]侯云鹤,熊信艮,吴耀武,等.基于广义蚁群算法的电力系统经济负荷分配[J].中国电机工程学报,2003,23(3):59-64.HouYunhe,XiongXinyin,Wu yaowu et al.Economic dispatch of power systems based on generalizedAnt colony optimization method[J].Proceedings of theCSEE,2003,23(3):59-64.
• [11]YuInkeun,ChouC S,SongY H.Application of the ant colony search algorithm to short-term generation scheduling problem of thermal units[C].Proceedings ofPOCON?8,InternationalConference onPowerSystemTechnology,Beijing,China,1998,552-556.
• [12]洪炳熔,金飞虎,高庆吉.基于蚁群算法的多层前馈神经网络[J].哈尔滨工业大学学报,2003,35(7):823-825.HongBingrong,Jin feihu,GaoQingji.Multi-layer feedforward neural network based on ant colony system[J].Journal ofHarbinInstitute ofTechnology,2003,35(7):823-825.
• [13]ZhangSubing,LiuZemin.NeuralNetworkTrainingUsingAntAlgorithmInATM TrafficControl[C].The2001IEEE InternationalSymposium on,2001,Vol.2,Page(s):157-160.
• [14]ShenghongLi,ZeminLiu.A GeneralCAC ApproachUsingNovelAntAlgorithmTrainingBasedNeuralNetwork[C].IJCNN '99.InternationalJointConference on,1999,Vol.3Page(s):1885-1888.
• [15]詹士昌,徐婕,吴俊.蚁群算法中有关算法参数的最优选择[J].科技通报,2003,19(5):381-386.ZhanShichang,XuJie,WuJun.TheOptimalSelection on theParameters of theAntColonyAlgorithm.Bulletin ofScience andTechnology[J].2003,19(5):381-386.
• [16]张智晟,孙雅明,何云鹏.基于相空间重构理论和优化递归神经网络结合的短期负荷预测方法[J].中国电力,2004,37(1):19-23.ZhangZhisheng,SunYaming,He yunpeng.A new approach ofSTLF based on combination of phase space reconstruction theory and optimal recursive neural networks[J].ElectricPower,2004,37(1):19-23.