摘要
饮用水安全事关国计民生,人类对水资源的过度开发导致的水环境问题对饮用水安全具有严重的威胁。水质预警系统对水质进行实时分析评价、预警,可以有效控制和减少水质恶化造成的危害,达到对水质恶化有效认知和控制的目的,使整个饮用水安全保障体系进入良性循环。水质预测是水质预警系统构建的一项基础性工作,及时有效的水质预测可以为水质预警系统提供可靠的评价及预警依据。
当前水质预测研究中,多采用单一水质预测方法对水质进行预测,某些针对特定单一预测方法的组合预测缺少一般框架性组合方法；概率性水质预测还没有引起广泛关注。针对这一状况,本文进行了基于概率组合的水质预测方法研究,并结合某水质预警项目,构建了典型渐变性水源水质污染预警系统。
本文的主要工作和创新点如下：
(1)提出了基于概率组合的水质预测方法。组合预测采用优势矩阵法对各单一预测方法加权融合,能够有效改善预测效果,并可以进一步扩展新方法；概率性预测基于对历史预测的统计,给出水质指标在一定置信度下的区间性预测结果。
(2)进行了旨在验证概率组合水质预测方法有效性的水质预测实验。利用基于支持向量回归机和灰色系统的水质预测模型以及二者的组合模型、基于BP神经网络和灰色系统的水质预测模型以及二者的组合模型对钱塘江某断面水质进行预测,并给出预测效果分析；进行概率性预测实验,并对概率性预测有效性进行检验。结果表明组合预测方法能够有效提高水质预测效果,对于不同的单一水质预测方法具有良好的适应性和自寻优能力；概率性预测有效性能够得到验证,并成功给出一定置信度下的区间性预测结果。
(3)构建了典型渐变性水源水质污染预警系统。基于概率组合水质预测方法,结合水质安全评价、预警信息发布等模块,进行了系统集成,建立了从现场水质监测信息到水质预测、水质安全评价、水质预警信息发布的典型渐变性水源水质污染预警系统,并结合水质预警课题实践,在三示范地进行了示范应用。
The safety of drinking water which has been threaten seriously by human over-exploitation of the water resources and environment is closely related to people's livelihood. Water quality early-warning system can effectively reduce and control the harm caused by the deterioration of water quality through the real time water quality monitor, assessment and early-warning. Prompt and effective water quality prediction can provide a reliable basis for water quality assessment and early-warning.
In the current study of water quality prediction, researchers prefer single prediction methods for water quality prediction, and the existing combination methods for water quality prediction lack of a framework approach. The probability prediction is always based on an assumption that water quality index follows certain probability distribution. Base on probability and combination, a method for water quality prediction is proposed in this thesis. A water quality early-warning system for typical sustained water pollution is built with the demands of the practice of a certain water quality early-warning project.
The main contents and innovative points are summarized as follows:
(1)A water quality prediction method based on probability and combination is proposed. The method combines the prediction results of different single methods through Odds-Matrix method and it can improve the performances of prediction effectively. It is worth noting that the combination-forecast approach can be extended to new methods. The probability of prediction is established through statistical analysis of historical prediction data and hence the validation of the method is achieved along with interval estimation under certain confidence level.
(2)A water quality prediction experiment aimed to verify the effectiveness of the method based on probability and combination is carried out. Two groups of water quality prediction methods are developed for a monitoring section of Qiantang River to conduct this work. One of the groups is the methods based on Grey Model theory. Support Vector Regression and the combination of them and the other one is the methods based on Grey Model theory, BP neural network and the combination of them. The analysis of the effects of these prediction methods and the experiment of probability prediction is also developed. The experimental results indicate that the combination-forecast approach performs better than single prediction methods. The validity of probability establishment can be checked effectively. According to the results, the interval prediction under certain confidence level can be given.
(3) A water quality early-warning system for typical sustained water pollution is built. The system integration is carried out with the combination of the modules of the water quality prediction method based on probability and combination, water quality safety assessment and early-warning information issue. And the early-warning system for typical sustained water pollution which can achieve a complete water quality early-warning process from the monitoring data to water quality prediction, safety assessment and information issue is established. With the practice of the water quality early-warning project, the system has been well applied in three demonstration areas.
引文
[1]樊敏,顾兆林.非机理性水质预测模型研究综述[J].环境科学与管理.2009,34(9)：63-67.
[2]EPA. Water Quality Models [EB/OL]. http://water.epa.gov/scitech/datait/models, 2012
[3]雒文生,宋星原.水环境分析及预测[M].武汉：武汉大学出版社,2000.
[4]谢永明.环境水质模型概论[M].西安：西安交通大学出版社,1987.
[5]田炜,王平,谢湉等.地表水质模型应用研究现状与趋势[J].现代农业科技.2008(3)：192-195.
[6]曹晓静,张航.地表水质模型研究综述[J].水利与建筑工程学报.2006,4(4)：18-21,52.
[7]蔡建安,张文艺,周志鹏等.Streeter-Phelps模型的缺陷及其改进研究[J].安徽工业大学学报.2003,20(4)：295-298,321.
[8]Wang P F, Martin J, Morrison G. Water Quality and Eutorphication in Tampa Bay,Florida[J]. Estuatine, Coastal and Shelf Science.1999, (49):1-20.
[9]Lewis D R, Willianms R J, Whitehead P G. Quality simulation along rivers(QUASAR):an application to the Yoikshire Ouse[J]. The science of Total Environment.194/195(1997):399-418.
[10]Kim T, Sheng Y P. Estimation of water quality model parameters [J]. Journal of Civil Engineering.2010,14(3):421-437.
[11]Sen P, George Y F, Zhao X. Integration of USEPA WASP model in a GIS platform[J]. I J Zhejiang Univ-Sci A.2010,11(12):1015-1024.
[12]Yuliana S, Tati T M, Jacub R, et al. Water Quality Modeling for Environmental Information System[C]. IEEE Asia-Pacific Conference on Circuits and Systems. 2004,2:929-932.
[13]Cox B A. A review of currently available in-stream water quality models and their applicability for simulating dissolved oxygen in lowland Rivers[J]. The Science of the Total Environment.2003,314(1):335-377.
[14]Streeter H W, Phelps E B. A study of the pollution and natural purification of the Ohio River[R]. United States Public Health Service Bulletin.1925,146.
[15]Chapra S C, Pelletier G J.QUAL2K:A modeling Framework for Simulating River and Stream Water Quality:Documentation and Users Manual[R]. Civil and Environmental Engineering Dept., Tufts University.2003.
[16]李云生,刘伟江,吴悦颖等.美国水质模型研究进展综述[J].水利水电技术.2006,37(2)：68-73.
[17]Park S S, Lee Y S. A water quality modeling study of the Nakdong River, Korea[J]. Ecological Modeling.2002,152:65-75.
[18]陈家军,于艳新,李森.QUAL2E模型在呼和浩特市水质模拟中的应用[J].水资源保护.2004,(3)：1-4,25,69.
[19]陈月,席北斗.何连生等.QUAL2K模型在西苕溪干流梅溪段水质模拟中的应用[J].环境工程学报,2008,2(7)：1000-1003.
[20]方晓波,张建英,陈伟等.基于QUAL2K模型的钱塘江流域安全纳污能力研 究[J].环境科学学报.2007,27(8)：1402-1407.
[21]陈美丹,姚琪,徐爱兰.WASP水质模型及其研究进展[J].水利科技与经济.2006,12(7)：420-426.
[22]Tim A W, Robert B A, Edword A C. Water quality analysis simulationprogram version 6.0 draft:user manual[R]. Region4, Atlanta, GA, USEPA, MS Tetre. Tech., Inc,2001.
[23]James T R, Martin J, Wool T, et al. A sediment resuspension and water quality model of Lake Okeechobee[J]. Journ.2011,23(4):107-109.
[24]刘兰岚,张永红.WASP水质模型在辽河干流污染减排模拟中的应用[J].环境科学与管理.2005,35(5)：160-163.
[25]Tufford D L, McKellar H N. Spatial and temporal hydrodynamicand water quality modeling analysis of a large reservoiron the South Carolina (USA)coastal plain[J]. Ecological Modelling.1999,114:137-173.
[26]贾海峰,程声通,杜文涛.GIS与地表水水质模型WASP5的集成[J].清华大学学报(自然科学版).2011,41(8)：125-128.
[27]Srinivasan R, Arnold J G. Intrgration of a basin-scale water quality model with GIS[J]. Journal of the American Water Resources Association.1994,30(3): 453-462.
[28]Udoyara S T, Robert J. Evaluating Agricultural Nonpoint-Source Pollution Using Integrated Geographic Information Systems and Hydrologic/Water Quality Model[J]. Journal of Environmental Quality.1993,23(1):25-35.
[29]Wang J P, Cheng S T, Jia H F. Water quality changing trends of the Miyun Reservoir[J]. Journal of Southeast University.2005,21(2):215-219.
[30]Liu X H, Huang W R. Modeling sediment resupension and transport induced by storm wind in Apalachicola Bay, USA[J].Environmental Modeling and Software. 2009,(24):1302-1313.
[31]Wool T A, Davie S R, Rodriguez. Development of three-dimensional hydrodynamic and water quality models to support total maximum daily load decision process for the Neuse River Estuary, North Carolina[J]. Journal of Water Resources Planning and Management.2003,139(4):295-306.
[32]Jeong S, Yeon K, Hur Y, et al. Salinity intrusion characteristics analysis using EFDC model in the downstream of Geum River[J]. Journal of Environmental Science.2010,22(6):934-939.
[33]史根香,郭海生.指数平滑法在地下水水质预测中的尝试[J].湖北地矿.1998,12(6)：35-40.
[34]吴涛,颜辉武,唐桂刚.三峡库区水质数据时间序列分析预测研究[J].武汉大学学报信息科学版.2006,31(6)：500-507.
[35]翟颢瑾,高晶.长江未来水质污染的时间序列分析[J].沈阳师范大学学报.2006,24(1)：22-24.
[36]Pavla P, Milan O, Jan P, et al. Prediction of water quality in the danube river under extreme hydrological and tempretature conditions[J]. J. Hydrol. Hydromech.2009,57(1):3-15.
[37]Nandish M M. Time series analysis of historical surface water quality data of the River Glen Catchment, U.K.[J]. Journal of Environmental.1996,46:149-172.
[38]Radwan M. Time Series Statistical Analysis of Water Quality Model Results for the Rosetta Branch of the Nile River [J]. Arab. Gulf. J. Of Scientific Reseatch. 2010,28(1):29-42.
[39]Awadallah A G, Hussam F, Karaman H G. Trend detection in water quality data using time series seasonal adjustment and statistical tests[J]. Irrig. and Drain.. 2011,60:253-262.
[40]牛军宜,冯平.基于Markov状态切换的水质时序自回归预测模型[J].吉林大学学报(地球科学版).2010,40(3)：657-664.
[41]颜剑波,阮晓红,孙瀚.多元回归分析在黄河水质预测中的应用[J].人民黄河.2010,32(3)：35-36.
[42]史复有,孙谦,李昌迪.黄河兰州段耗氧有机污染物浓度统计预测模型的建立[J].环境科学.1989,16(3)：72-74,82.
[43]向速林.基于回归分析的地下水水质预测研究[J].东华理工学院学报.2007,30(2)：161-163.
[44]Liu Y, Guo H C, Yang P J. Exploring the influence of lake water chemistry on chlorophylla: a multivariate statistical model analysis[J]. Ecological Modelling. 2010,221(4):681-688.
[45]李栋臣.逐步回归分析在地下水水质预测中的应用[J].焦作矿业学院学报.1992,29(4)：8-15.
[46]李如忠.水质预测理论模式研究进展与趋势分析[J].合肥工业大学学报(自然科学版).2006,29(1)：26-30.
[47]邓聚龙.灰色预测与决策[M].武汉：华中理工大学出版社.1986：75-76.
[48]孙志霞,孙英兰.GM(1,1)模型研究及其在水质预测中的应用[J].海洋通报,2009,28(4)：116-120.
[49]王泽斌,马云,叶珍等.应用GM(1,1)模型预测阿什河水质变化趋势[J].环境科学与管理.2011,36(4)：24-27,39.
[50]李如忠,汪家权,钱家忠.基于灰色动态模型群法的河流水质预测研究[J].水土保持通报,2002,22(4)：10-12.
[51]马防.新陈代谢GM(1,1)模型在河流水质预测中的应用[J].山西建筑.2008,34(16)：169-170.
[52]王开章,刘福胜,孙鸣.灰色模型在大武水源地水质预测中的应用[J].山东农业大学学报(自然科学版).2002,33(1)：66-71.
[53]Holger R M, Graeme C D. Neural networks for the prediction and forecasting of water resources variables:a review of modelling issues and applications [J]. Environmental Modelling and Software.2000, (15):101-124.
[54]李成林,李鸿雁,鲍新华等.学习率有限监督调整BP网络在黄河下游水质预测中的应用[J].东北水利水电.2008,26(293)：5-6.
[55]郭劲松,霍国友,龙腾锐.BOD-DO耦合人工神经网络水质模拟的研究[J].环境科学学报.2001,21(2)：140-143.
[56]牟洁.基于神经网络和灰色系统的水质预测研究[J].天津：天津大学,2010,
[57]过仲阳,陈中原,李绿芊等.人工神经网络技术在水质动态预测中的应用[J].华东师范大学学报(自然科学版).2001,(1)：84-89.
[58]梁楠.基于人工神经网络的水质预测及Matlab实现[硕士学位论文].陕西,长安大学,2007.
[59]莫慧芳,谷爱昱,张新政.基于BP神经网络的水质预测方法的研究[J].控制工程.2004,11(增刊)：9-10,19.
[60]王晓萍,孙继洋,金鑫.基于BP神经网络的钱塘江水质指标的预测[J].浙江大学学报(工学版).2007,41(2)：361-364.
[61]Gunn S R. Support Vector Machines for Classification and Regression[R].University of Southhampton,1998.
[62]Boser B E, Guyon I M, Vapnik V N. A training algorithm for optimal margin classifiers[C]. New York:ACM,1992.
[63]Bouamar M, Ladjal M. Evaluation of the performances of ANN and SVM techniques used in water quality classification[C]. Marrakech:14th IEEE International conference on Electronics, Circuits and Systems,2007.
[64]Mohamed B, Mohamed L. A comparative study of RBF neural network and SVM classification techniques performaned on real data for drinking water quality[C]. 5th International Multi-Conference on System, Signals and Device,2008.
[65]Vapnik V, Golowich S E, Smola A. Support Vector Method for Function Approximation, Regression Estimation, and Signal Processing[C]. Advances in Neual Information Processing Systems,1996.
[66]Muller K R, Smola A J, Ratsch G et al. Predicting time series with support vector machines[C]. Artificial Neural Networks-ICANN Lecture Notes in Computer Science.1997,1327:999-1004.
[67]Xiang Y R, Jiang L Z. Water Quality prediction using LS-SVM with particle swarm optimization[C]. Second International Workshop on Knowledge Discovery and Data Mining,2009.
[68]徐红敏.王继广.加权支持向量机及其在水质预测中的应用[J].世界地质.2007,26(1)：58-61.
[69]梁雪春,龚艳冰,肖迪.一种多核加权支持向量机的水质预测方法[J].东南大学学报(自然科学版).2011,41(增刊)：14-17.
[70]房平,邵瑞华.司全印等.最小二乘支持向量机应用于西安灞河口水质预测[J].系统工程.2011,29(6)：113-117.
[71]Cao J. Hu H S, Qian S X, et al. Research on the water quality forecast method based on SVM[C]. Proc. SPIE 7500,750005,2009.
[72]石为人,王燕霞,唐云建等.小样本跳变水质时序数据预测方法[J].计算机应用.2010,30(2)：486-489,505.
[73]Bates J M, Granger. The combination of Forecasts[J]. Operational Research Society.1969,20(4):451-468.
[74]唐小我,曹长修,金德运.组合预测最有加权系数向量的进一步研究[J].预测.1994,(2)：48-49.
[75]汪同三,张涛.组合预测—理论、方法及应用[M].北京：社会科学文献出版社.2008.
[76]Gupta S, Wilton P C. Combination of forecasts:an extension[J].Management Science,1987,33(3):356-372.
[77]谢开贵,周家启.变权组合预测模型研究[J].系统工程理论与实践.2000,(7)：36-40,117.
[78]傅庚,唐小我,曾勇.广义递归方差倒数组合预测方法研究[J].电子科技大 学学报.1995,24(2)：211-217.
[79]陈华友,侯定丕.基于预测有效度的组合预测模型研究[J].预测.2001,20(3)：72-73.
[80]Swanson N R, Zeng T. Choosing among competing econometric forecasts: Regression-based forecast combination using model selection] [J]. Journal of Forecasting.2001,20(6):425-440.
[81]Stock J H, Watson M W. Combination forecasts of output growth in a seven-country data set[J]. Journal of Forecasting.2004,23(6):405-430.
[82]Clemen R T, Winkler R L. Combining Economic Forecasts[J]. Journal of Business and Economic Statistics.1986,4(1):39-46.
[83]Lawrence M J, Edmundson R H, O'Connor M J. The accuracy of combining judgement and statistical forecasts[J]. Management Science.1986,32(12): 1521-1532.
[84]Andrawis R R, Atiya A F. EI-Shishiny H. Combination of long term and short term forecasts, with application to tourism demand forecasting [J]. International Journal of Forecasting.2011.27(3):870-886.
[85]尚晓锶,林卫东,唐艳葵.指数平滑和GM(1,1)组合法在水质预测中的应用—以邕江水源地铁、锰浓度为例[J].环境科学与技术.2011,34(1)：192-195.
[86]Faruk D O. A hybrid neural network and ARIMA model for water quality time series prediction[J]. Engineering Applications of Artificial Intelligence,2010,23(4):586-594.
[87]Zhang G P. Time series forecasting using a hybrid ARIMA and neural network model[J]. Neurocomputing.2003,50:159-175.
[88]牛志广,张宏伟,辛志伟.基于log-logistic概率分布的近海水质组合预测方法研究[J].2006,26(5)：111-116.
[89]姜云超,南忠仁.不确定性水质模型及其研究进展[J].环境污染与防治.2007,29(9)：713-717.
[90]Takyi A K. Surface water quality management using a multiple-realization chance constraint method[J]. Water Resources Research.1999,35(5):1657-1670.
[91]Beck M B. Water quality modeling:a review of the analysis of uncertainty [J]. WaterResources Research,1987(23):1393-1442.
[92]Bhore R N. Uncertainty analysis in large models[phD theis]. University of Delaware.1996.
[93]高顺清,白仁海.降水概率预报综述[J].黑龙江气象.1996,(3)：1-2.
[94]杨文佳,康重庆,夏清,等.基于预测误差分布特性统计分析的概率性短期负荷预测[J].电力系统自动化,2006,30(19)：47-52.
[95]Wendroth O, Rogasik H. Koszinski S, et al.state space prediction of field scale soil water content time series in a study loam[J]. Sail and Tillage Research, 1999.50(1):85-93.
[96]Finney B A, Bowles D S, Windlham M P. Random differential equations in river water quality modeling[J]. Water Resources Research.1982,18(1):122-134.
[97]Marier H R, Morgan N, Chow C W K. Use of artificial neural network for predicting optimal alum doses and treated water quality parameters [J]. Environmental Modeling and Software.2004,19(5):485-494.
[98]Ghosh S, Mujumdar P P. Risk minimization in water quality contol problems of a river system[J]. Advanced in water resources.2006,29(3):458-470.
[99]Karmakara S, Mujumdar P P. Grey fuzzy optimization model for water quality management of a river system[J]. Advances in Water Resource.2006,29(7): 1088-1105.
[100]Park J-Il, Jung N-C, Kwak K-C. Water quality prediction in a reservoir: linguistic model approach for interval prediction[J]. International Journal of Control Automation and Systems,2010,8(4):868-874.
[101]陈华友.组合预测方法有效性理论及其应用[M].北京：科学出版社,2008.
[102]李庆扬,王能超,易大义.数值分析[M].北京清华大学出版社,2002.
[103]盛骤,谢式千,潘承毅.概率论与数理统计[M].北京：高等教育出版社,2008.
