灌区年径流量预测与灌区优化配水研究
|
摘要
随着社会经济的发展,水资源短缺问题日益严重,因此合理配置与有效管理水资源是灌区实现节水与取得效益的重要保证。本文以年径流量预测与灌区优化配水为研究目标,主要进行的工作及技术实现包括以下三个方面:
首先,在对灌区年径流量相关概念与影响因素研究的基础上,分析了年径流量具有的特点,同时比较分析了年径流量的预测方法,针对年径流量具有的特点,选取适宜的预测方法,确定采用支持向量机方法进行年径流量预测;
其次,年径流量的变化情势受气候、下垫面、太阳活动与人类活动等因素的影响,这些影响因素的复杂性与不确定性,给年径流量预测带来了极大的难度,基于支持向量机方法能够较好地解决小样本、高维、非线性问题,建立年径流量SVM预测模型,所建立的模型预测结果精度高,证明了该方法应用的合理性和有效性;
最后,针对灌区水资源不足时的非充分灌溉情况,为了使灌区取得最大的灌溉经济效益,建立灌区优化配水模型,利用遗传算法对所建立的优化模型求解,结果显示采用遗传算法可以获得较好的寻优效果,有效提高了灌区灌溉经济效益。
With the development of economic society,the problem of water shortage is increasingly serious,therefore,the rational allocation and effective management of the water resources is the important guarantee for saving water resources and making the benefit.This paper takes annual runoff prediction and an optimal allocation of water as research aims,the work and technical implementation including the following three aspects:
First,on the basis of the research on the irrigation area in annual runoff and influencing factors,analyzes the characteristics of the annual runoff,and compara the time rank,artificial neural network and support vector machine.For the characterisics of the annual runoff,use the support vector machine to forecast the annual runoff.
Secondly,the change of annual runoff is influenced on some complicated characteristics,such as climate,surface,solor activity and human activity.The complexity and uncertainty of these factors bring great difficulties to the annual runoff.Based on support vector machine can be used to solve the small sample,high dimension,nonlinear problems,set up a support vector machine prediction model.The SVM prediction model set up achieves a better prediction result.
Last,when the water resoures is insufficient in the irrigation areas,in order to achieve the greatest agricultural ecnomical benefit,set up an optimal allocation model of water,Using the genetic algorithm to solve the established model solution,it is indicated that genetic algorithm can gain a better effect of the optimization,and the efficiencies of water resources utilization are improved.
首先,在对灌区年径流量相关概念与影响因素研究的基础上,分析了年径流量具有的特点,同时比较分析了年径流量的预测方法,针对年径流量具有的特点,选取适宜的预测方法,确定采用支持向量机方法进行年径流量预测;
其次,年径流量的变化情势受气候、下垫面、太阳活动与人类活动等因素的影响,这些影响因素的复杂性与不确定性,给年径流量预测带来了极大的难度,基于支持向量机方法能够较好地解决小样本、高维、非线性问题,建立年径流量SVM预测模型,所建立的模型预测结果精度高,证明了该方法应用的合理性和有效性;
最后,针对灌区水资源不足时的非充分灌溉情况,为了使灌区取得最大的灌溉经济效益,建立灌区优化配水模型,利用遗传算法对所建立的优化模型求解,结果显示采用遗传算法可以获得较好的寻优效果,有效提高了灌区灌溉经济效益。
With the development of economic society,the problem of water shortage is increasingly serious,therefore,the rational allocation and effective management of the water resources is the important guarantee for saving water resources and making the benefit.This paper takes annual runoff prediction and an optimal allocation of water as research aims,the work and technical implementation including the following three aspects:
First,on the basis of the research on the irrigation area in annual runoff and influencing factors,analyzes the characteristics of the annual runoff,and compara the time rank,artificial neural network and support vector machine.For the characterisics of the annual runoff,use the support vector machine to forecast the annual runoff.
Secondly,the change of annual runoff is influenced on some complicated characteristics,such as climate,surface,solor activity and human activity.The complexity and uncertainty of these factors bring great difficulties to the annual runoff.Based on support vector machine can be used to solve the small sample,high dimension,nonlinear problems,set up a support vector machine prediction model.The SVM prediction model set up achieves a better prediction result.
Last,when the water resoures is insufficient in the irrigation areas,in order to achieve the greatest agricultural ecnomical benefit,set up an optimal allocation model of water,Using the genetic algorithm to solve the established model solution,it is indicated that genetic algorithm can gain a better effect of the optimization,and the efficiencies of water resources utilization are improved.
引文
[1]杨旭,栾继虹,冯国璋.中长期水文预报研究评述与展望[J].西北农业大学学报,2000,28(6):203-207.
[2]顾绮仪.长期水文统计预报方法在长江大通站径流预报中的应用[J].水利水文自动化,2005,3(1):42-43.
[3]陈守煜.模糊水文学[J].大连理工大学学报,1988(1):93-97.
[4]陈守煜.模糊水文学与水资源系统模糊优化原理[M].大连:大连理工出版社,1990.
[5]王本德,曹玉涛,唐国磊.于桥水库来水量模糊推理预报及引水方式研究[J].人民黄河,2007.29(1):63-63.
[6]ASCE Task Committee.Artificial neural networks in hydrology Ⅰ:Preliminary concepts[J].Journal of hydrologic Engineering,2000,5(2):115-123.
[7]ASCE Task Committee.Artificial neural networks in hydrology Ⅱ:Hydrological applications[J].Journal of Hydrologic Engineering,2000,5(2):124-137.
[8]薛桂先.神经网络方法在黄河来水预测中的应用[J].农业网络信息,2006,(5),36-38.
[9]雷晓云,张丽霞,梁新平.基于MATLAB工具箱的BP神经网络年径流量预测模型研究[J].水文,2008,28(1),43-46.
[10吴超宇.张文.水文预报的人工神绎网络方法[J].中山大学学报(自然科学版),1994.33(1),79-90.
[11]何万科.基于灰色系统理论的千河枯水年预测[J].水资源与水工程学报,2008,19(2),41-43.
[12]夏军.中长期径流预估的一种灰关联模式与预测方法[J].水科学进展,1993,4(3):190-197.
[13]冯平,杨鹏,李润苗.枯水期径流量的中长期预报模式[J].水利水电技术,1992,23(2):6-9.
[14]Barrett.J.W.H.,Skogerboe,G.V.,Crop production funcations and the allocation and use of irrigation water.Agricultural Water Management,Vol.3,No.1,1980.
[15]高正廉.冬小麦灌溉方案优化的探讨[J].灌溉排水,1988,2.
[16]孙顺松.灌区经济效益分析[J].西北水利科技,1988,1.
[17]康绍忠,党育红.作物水分生产函数与经济用水灌概制度的研究[J].西北水利科技,1987年第1期.
[18]陈守煜.复杂水资源系统优化模糊识别理论与应用[M].吉林:吉林大学出版,社,2002.
[19]王德智.中长期水文预报的系统理论与方法研究[D].武汉:武汉大学 2004.
[20]Tucci C E M,Clarke R T,CollischomW,et al.Long-term flow forecasts based climate and hydrologic modeling:Uruguay River basin[J].Water Resoures,2003,39(7).
[21]Piechota T C,Dracup J A.Long-range stream-flow forecasting using El-Nino southern oscillation indications[J].Journal of Hydrologic Enginee ing,1999,4(2):144-151.
[22]刘少华,毛红梅.一种水文时间序列预报的新方法[J].武汉大学学报(工学版),2002,35(4):52-55.
[23]丁晶,邓育仁.随机水文学[M].成都:成都科技大学出版社,1988年10月.
[24]顾正华,唐洪武等.水流模拟智能化问题的探讨[J].水科学进展,2004,15(1):129-133.
[25]黄忠恕,王钦梁,匡奇.北太平洋和青藏高原下垫面热状况与长江流域汛期旱涝关系初步探讨.水文预报论文选集(1981年全国水文预报学术讨论会)[J],北京,1985.
[26]刘清仁.松花江流域水旱灾害发生规律及长期预报研究[J].水科学进展,1994,5(4):319-327.
[27]章淹.致洪暴雨中期预报进展[J].水科学进展,1995,6(2):162-168.
[28]焦李成.神经网络系统理论[M].西安:西安电子科技大学出版社.1991.
[29]VapnikV N.The Nature of Statistical Learning Theory.NY:Sprionger-Verlag,1995.
[30]Vladimir Cherkassky and Yunqian Ma,Practical Selection of SVM Parameters and Noise Estimation for SVM Regression,2004
[31]Gary William Flake,Steve Lawrence.Efficient SVM Regression Training with SMO.Machine Learning,2001.
[32]张学工.关于统计学理论与支持向量机[J].自动化学报,2001,26(1).
[33]王国胜,钟义信.支持向量机的若干新进展[J].电子学报,2001,1:58-61.
[34]http://www.csie.ntu.edu.tw/～cjlin/
[35]刘国东,丁晶.BP网络用于水文预测的几个问题探讨[J].水利学报,1999,(1):65-69.
[36]Hsu K,Gupta H V and Sorooshian S.Artifical network modeling of the rainfall-runoff process.Water Resources Research,1995,31(10):2517-2530.
[37]Raman H and Sunilkumar N.Multivarate modeling of water resources time series using artificial neural networks.Hydrological Sciences Journal,1995,40(2):145-163.
[38]中华人民共和国水利部.水文情报预报规范(SL250-2000).
[39]胡敬鹏,周新志.都江堰灌区渠首优化配水模型探讨[J].水利科技与经济,2006,12(1):55-57.
[40]陈文伟,黄金才,赵新昱著.数据挖掘技术[M].北京:北京工业大学出版社,2002.
[41]王小平、曹立明著.遗传算法--理论、应用与软件实现[M].西安:西安交通大学出版社,2002.
[42]金菊良,丁晶.水资源系统工程[M].成都:四川科学技术出版社,2002.
[43]丁晶,金菊良,杨晓华等.基因算法在水科学中的应用[J].人民长江,1999,30(增刊):13-15.
[44]WARDLAW R,SHARIF M.Evaluation of genetic algorithms for optimal reservoir resources[J].Journal of Water Resource Planning and Managemen,1999,125(1):25-33.
[45]SHARIF M,WARDLAW R.Multireservoir systems optimization using genetic algorithms:case study[J].Journal of Computing in Civil Engineering,2000,14(4):255 -263.
[46]CHANG Li-chiu,CHANG Fi-john.Intelligent control for modelling of real time reservoir operation[J].Hydrological Processes,2001,15(9):1621-1634.
[47]HUANG Wen-cheng,YUAN Lun-chin,CHI Ming.Linking genetic algorithms with stochastic dynamic programming to the long 2term operation of a multireservoir system [J].Water Resources Research,2002,38(12),1304,doi:1011029/2001 WR 0011 22.
[2]顾绮仪.长期水文统计预报方法在长江大通站径流预报中的应用[J].水利水文自动化,2005,3(1):42-43.
[3]陈守煜.模糊水文学[J].大连理工大学学报,1988(1):93-97.
[4]陈守煜.模糊水文学与水资源系统模糊优化原理[M].大连:大连理工出版社,1990.
[5]王本德,曹玉涛,唐国磊.于桥水库来水量模糊推理预报及引水方式研究[J].人民黄河,2007.29(1):63-63.
[6]ASCE Task Committee.Artificial neural networks in hydrology Ⅰ:Preliminary concepts[J].Journal of hydrologic Engineering,2000,5(2):115-123.
[7]ASCE Task Committee.Artificial neural networks in hydrology Ⅱ:Hydrological applications[J].Journal of Hydrologic Engineering,2000,5(2):124-137.
[8]薛桂先.神经网络方法在黄河来水预测中的应用[J].农业网络信息,2006,(5),36-38.
[9]雷晓云,张丽霞,梁新平.基于MATLAB工具箱的BP神经网络年径流量预测模型研究[J].水文,2008,28(1),43-46.
[10吴超宇.张文.水文预报的人工神绎网络方法[J].中山大学学报(自然科学版),1994.33(1),79-90.
[11]何万科.基于灰色系统理论的千河枯水年预测[J].水资源与水工程学报,2008,19(2),41-43.
[12]夏军.中长期径流预估的一种灰关联模式与预测方法[J].水科学进展,1993,4(3):190-197.
[13]冯平,杨鹏,李润苗.枯水期径流量的中长期预报模式[J].水利水电技术,1992,23(2):6-9.
[14]Barrett.J.W.H.,Skogerboe,G.V.,Crop production funcations and the allocation and use of irrigation water.Agricultural Water Management,Vol.3,No.1,1980.
[15]高正廉.冬小麦灌溉方案优化的探讨[J].灌溉排水,1988,2.
[16]孙顺松.灌区经济效益分析[J].西北水利科技,1988,1.
[17]康绍忠,党育红.作物水分生产函数与经济用水灌概制度的研究[J].西北水利科技,1987年第1期.
[18]陈守煜.复杂水资源系统优化模糊识别理论与应用[M].吉林:吉林大学出版,社,2002.
[19]王德智.中长期水文预报的系统理论与方法研究[D].武汉:武汉大学 2004.
[20]Tucci C E M,Clarke R T,CollischomW,et al.Long-term flow forecasts based climate and hydrologic modeling:Uruguay River basin[J].Water Resoures,2003,39(7).
[21]Piechota T C,Dracup J A.Long-range stream-flow forecasting using El-Nino southern oscillation indications[J].Journal of Hydrologic Enginee ing,1999,4(2):144-151.
[22]刘少华,毛红梅.一种水文时间序列预报的新方法[J].武汉大学学报(工学版),2002,35(4):52-55.
[23]丁晶,邓育仁.随机水文学[M].成都:成都科技大学出版社,1988年10月.
[24]顾正华,唐洪武等.水流模拟智能化问题的探讨[J].水科学进展,2004,15(1):129-133.
[25]黄忠恕,王钦梁,匡奇.北太平洋和青藏高原下垫面热状况与长江流域汛期旱涝关系初步探讨.水文预报论文选集(1981年全国水文预报学术讨论会)[J],北京,1985.
[26]刘清仁.松花江流域水旱灾害发生规律及长期预报研究[J].水科学进展,1994,5(4):319-327.
[27]章淹.致洪暴雨中期预报进展[J].水科学进展,1995,6(2):162-168.
[28]焦李成.神经网络系统理论[M].西安:西安电子科技大学出版社.1991.
[29]VapnikV N.The Nature of Statistical Learning Theory.NY:Sprionger-Verlag,1995.
[30]Vladimir Cherkassky and Yunqian Ma,Practical Selection of SVM Parameters and Noise Estimation for SVM Regression,2004
[31]Gary William Flake,Steve Lawrence.Efficient SVM Regression Training with SMO.Machine Learning,2001.
[32]张学工.关于统计学理论与支持向量机[J].自动化学报,2001,26(1).
[33]王国胜,钟义信.支持向量机的若干新进展[J].电子学报,2001,1:58-61.
[34]http://www.csie.ntu.edu.tw/～cjlin/
[35]刘国东,丁晶.BP网络用于水文预测的几个问题探讨[J].水利学报,1999,(1):65-69.
[36]Hsu K,Gupta H V and Sorooshian S.Artifical network modeling of the rainfall-runoff process.Water Resources Research,1995,31(10):2517-2530.
[37]Raman H and Sunilkumar N.Multivarate modeling of water resources time series using artificial neural networks.Hydrological Sciences Journal,1995,40(2):145-163.
[38]中华人民共和国水利部.水文情报预报规范(SL250-2000).
[39]胡敬鹏,周新志.都江堰灌区渠首优化配水模型探讨[J].水利科技与经济,2006,12(1):55-57.
[40]陈文伟,黄金才,赵新昱著.数据挖掘技术[M].北京:北京工业大学出版社,2002.
[41]王小平、曹立明著.遗传算法--理论、应用与软件实现[M].西安:西安交通大学出版社,2002.
[42]金菊良,丁晶.水资源系统工程[M].成都:四川科学技术出版社,2002.
[43]丁晶,金菊良,杨晓华等.基因算法在水科学中的应用[J].人民长江,1999,30(增刊):13-15.
[44]WARDLAW R,SHARIF M.Evaluation of genetic algorithms for optimal reservoir resources[J].Journal of Water Resource Planning and Managemen,1999,125(1):25-33.
[45]SHARIF M,WARDLAW R.Multireservoir systems optimization using genetic algorithms:case study[J].Journal of Computing in Civil Engineering,2000,14(4):255 -263.
[46]CHANG Li-chiu,CHANG Fi-john.Intelligent control for modelling of real time reservoir operation[J].Hydrological Processes,2001,15(9):1621-1634.
[47]HUANG Wen-cheng,YUAN Lun-chin,CHI Ming.Linking genetic algorithms with stochastic dynamic programming to the long 2term operation of a multireservoir system [J].Water Resources Research,2002,38(12),1304,doi:1011029/2001 WR 0011 22.