基于贝叶斯方法的三峡库区水环境质量研究
|
摘要
长江三峡工程是举世瞩目的跨世纪特大型水利工程,在保障防洪安全和综合开发利用方面具有举足轻重的作用。工程的建设对生态与环境的影响一直是国内外专家学者广泛关注的热点之一,因此本文对广泛关注的三峡库区水环境质量状况进行分析研究。
本论文主要以三峡蓄水运行3年多的水文水质监测资料为基础,从135m蓄水前后水质参数变化趋势,156m蓄水对坝下游水质影响及蓄水期间干支流水质变化几个方面展开分析。结果表明:蓄水以后水质整体水平呈下降趋势,但短期内没有明显恶化。支流水体蓄水后,水体营养盐浓度受干流影响增加,在其他条件适合的情况下,会加速水体的富营养化。
在比较众多评价方法的基础上,论文重点研究建立了贝叶斯正则化神经网络(BRNN)水质评价模型和贝叶斯水质评价模型。为避免因样本资料过少而使网络陷入局部最小值,从而引入贝叶斯规则化方法改进网络的自适应能力,从而改进了BRNN水质评价模型,同时本文尝试采用抽样误差正态分布原理来估计水质评价的似然概率,从而改进贝叶斯水质评价方法,使其更加合理和可靠。通过实例的评价对比分析,验证了模型的可靠性、合理性与实用性。从而合理评价了三峡库区水环境状况,客观地反映了三峡水库库区水环境特征与实际现状,为三峡水库水资源保护和监督管理提供科学依据和技术支持。
运用BRNN和Bayes方法分别从综合因子及分类因子(氧平衡因子、营养盐类因子、毒物因子)方面对三峡库区水质进行评价。结果表明:三峡库区总体水质良好,水质保持在Ⅰ、Ⅱ类水平,满足库区水体功能要求。建库蓄水后,局部水域的富营养化可能引起水华等环境灾害的发生;坝下水中溶解氧浓度出现超饱和现象,不利于鱼类的生长,应引起重视。
The Three Georges Project is the key project for yangtze river development. The impacts of project construction on ecological and environment were always the concern of domestic and foreign researchers,so this paper analyzes and studys the water quality conditions of the Three Georges Reservoir with wide attention.
On the basis of Three Gorges water running more than three years of water quality monitoring data, This paper mainly from 135m water quality parameters before and after the change trend, 156m water storage impacting on dam downstream water quality and the changes of mainstream and tributaries analysis the changes in several aspects of water quality. The results show that: the overall level of water quality is a declining trend, but in the short term there is no obvious deterioration of water quality. After tributaries storage, water nutrient concentrations affected by the increase in the mainstream, in other conditions suitable circumstances, the water body will be accelerated with the rich of nutrition.
On the basis of comparing various evaluation methods ,the paper focus on the establishment of regular Bayesian neural network (BRNN) water quality evaluation model and Bayesian water quality evaluation model. To avoid samples because of too little information getting into network a local minimum, this paper introduces Bayesian methods to improve the rules of the adaptive ability of the network, thus improving the water quality evaluation model BRNN, At the same time sampling error in this paper to try to estimate the normal principles of the water quality evaluation likelihood probability, thereby improving water quality evaluation Bayesian methods to make it more reasonable and reliable.Reliability,rationality and practicability of the model was verified by comparing analysis.The water environment status close to Three Georges Dam is evaluated correctly. In addition scientific and technical supprots are supplied for water quality protection and supervision of The Three Georges reservoir.
The water quality assessment on comprehensive factor and classify factor (oxygen balance factor, nutrients factor and toxic factor) in the Three Georges area was well generally and was not getting poor obviously after impounding water using BRNN models and Bayes model. Variation tendency of water quality during deposit months isⅠ、Ⅱlevels. Eutrophication in some local water area after impounding water may induce algae blooms. DO level in downstream of the dam is oversaturated not conducive to the growth of fish ,it should be regarded.
本论文主要以三峡蓄水运行3年多的水文水质监测资料为基础,从135m蓄水前后水质参数变化趋势,156m蓄水对坝下游水质影响及蓄水期间干支流水质变化几个方面展开分析。结果表明:蓄水以后水质整体水平呈下降趋势,但短期内没有明显恶化。支流水体蓄水后,水体营养盐浓度受干流影响增加,在其他条件适合的情况下,会加速水体的富营养化。
在比较众多评价方法的基础上,论文重点研究建立了贝叶斯正则化神经网络(BRNN)水质评价模型和贝叶斯水质评价模型。为避免因样本资料过少而使网络陷入局部最小值,从而引入贝叶斯规则化方法改进网络的自适应能力,从而改进了BRNN水质评价模型,同时本文尝试采用抽样误差正态分布原理来估计水质评价的似然概率,从而改进贝叶斯水质评价方法,使其更加合理和可靠。通过实例的评价对比分析,验证了模型的可靠性、合理性与实用性。从而合理评价了三峡库区水环境状况,客观地反映了三峡水库库区水环境特征与实际现状,为三峡水库水资源保护和监督管理提供科学依据和技术支持。
运用BRNN和Bayes方法分别从综合因子及分类因子(氧平衡因子、营养盐类因子、毒物因子)方面对三峡库区水质进行评价。结果表明:三峡库区总体水质良好,水质保持在Ⅰ、Ⅱ类水平,满足库区水体功能要求。建库蓄水后,局部水域的富营养化可能引起水华等环境灾害的发生;坝下水中溶解氧浓度出现超饱和现象,不利于鱼类的生长,应引起重视。
The Three Georges Project is the key project for yangtze river development. The impacts of project construction on ecological and environment were always the concern of domestic and foreign researchers,so this paper analyzes and studys the water quality conditions of the Three Georges Reservoir with wide attention.
On the basis of Three Gorges water running more than three years of water quality monitoring data, This paper mainly from 135m water quality parameters before and after the change trend, 156m water storage impacting on dam downstream water quality and the changes of mainstream and tributaries analysis the changes in several aspects of water quality. The results show that: the overall level of water quality is a declining trend, but in the short term there is no obvious deterioration of water quality. After tributaries storage, water nutrient concentrations affected by the increase in the mainstream, in other conditions suitable circumstances, the water body will be accelerated with the rich of nutrition.
On the basis of comparing various evaluation methods ,the paper focus on the establishment of regular Bayesian neural network (BRNN) water quality evaluation model and Bayesian water quality evaluation model. To avoid samples because of too little information getting into network a local minimum, this paper introduces Bayesian methods to improve the rules of the adaptive ability of the network, thus improving the water quality evaluation model BRNN, At the same time sampling error in this paper to try to estimate the normal principles of the water quality evaluation likelihood probability, thereby improving water quality evaluation Bayesian methods to make it more reasonable and reliable.Reliability,rationality and practicability of the model was verified by comparing analysis.The water environment status close to Three Georges Dam is evaluated correctly. In addition scientific and technical supprots are supplied for water quality protection and supervision of The Three Georges reservoir.
The water quality assessment on comprehensive factor and classify factor (oxygen balance factor, nutrients factor and toxic factor) in the Three Georges area was well generally and was not getting poor obviously after impounding water using BRNN models and Bayes model. Variation tendency of water quality during deposit months isⅠ、Ⅱlevels. Eutrophication in some local water area after impounding water may induce algae blooms. DO level in downstream of the dam is oversaturated not conducive to the growth of fish ,it should be regarded.
引文
[1]联合国环境规划署.全球环境展望2000[M].北京:中国环境科学出版社,2001.
[2]国家环境保护总局,2000年中国环境状况公报[J].2000.
[3]高千红.概率神经网络在三峡近坝水域水质评价中的应用研究[D].南京:河海大学,2006.
[4]马倩如,程声通等著.环境质量评价[M].北京:中国环境科学出版社,1990.
[5]环境质量评价原理与方法[M].北京:化学工业出版社,2004.
[6]Horton,R.K.,"An Index-Number System for Rating Water Quality,"Journal of Water Pollution Control Federation,37(3),300-305(1965).
[7]Mance,Elisabeth,An Inexact Two-stage Water Quality Management Model for the He River Basin in Yongxin County,MASc thesis,Jun.2007.
[8]Brown,R.M.,McClelland,N.I.,Deininger,R.A.,Tozer,R.G.,"A Water Quality Index- Do We Dare? ",Water Sewage Works.,117,3139-343(1970).
[9]Harkins,R.S.,"An objective water quality index,"Journal of Water Pollution Control Federation,46(3),588-591(1974).
[10]S.L.Ross.An index system for classifying river water quality.Water Pollut Control(1977),113-122.
[11]张炳文,韩晓滨,杨茂,闫琳.模糊综合指数法在呼和浩特市地面水环境质量综合评价中的应用[J].干旱区资源与环境,1996,10(03):87-92..
[12]李伯昌,施慧燕.长江口河段水环境现状分析[J].水资源保护,2005,21(01):39-44.
[13]沈新强,晁敏.长江口及邻近渔业水域生态环境质量综合评价[J].农业环境科学学报,2005,24(02):270-273.
[14]李柞泳,郭丽婷,欧阳洁.水环境质量评价的普适指数公式[J].环境科学研究,2001,14(3):56-58.
[15]刘征,刘洋.水污染指数评价方法与应用分析[J].南水北调与水利科技,2005.8,3(4).
[16]史宝忠.水质评价的一种新方法一水质综合特征模式[J].中国环境科 学,1994,14(4).
[17]翟由涛,赵玉军.模糊综合指数在判断水质变化趋势和水体管理中的应用[J].环境工程,1995.12,13(6).
[18]陈运东.模糊综合评判法在水质评价中的应用[J].四川水利,1999,20(02):56-58.
[19]徐祖信.我国河流单因子水质标识指数评价方法研究[J]同济大学学报(自然科学版),2005,33(03):321-325.
[20]马成有,曹剑锋,姜纪沂,平建华.改进的尼梅罗污染指数法及其应用-以磐石市地下水环境质量评价为例[J].水资源保护,2006,22(04):53-55.
[21]梁德华.河流水质综合评价方法的统一和改进[J].中国环境监测,2002.18(2):63-66.
[22]邹海明,李粉茹.基于实例的水质模糊综合评价[J].安徽农学通报,2005.11(4):154-155.
[23]郭劲松,龙腾锐等.四种水质综合评价方法的比较[J].重庆建筑大学学报,2000.22(4):6-12.
[24]韩国新,刘明华等.关于区域环境质量的复合模糊距离评价方法及应用探讨[J].中国环境监测,1996,12(6):35-39.
[25]梅学彬,王福纲.模糊综合评价法在水质评价中应用与探讨[J].世界地质,2000.19(2):172-177.
[26]曾永,樊引琴等.水质模糊综合评价法与单因子指数评价法比较[J].人民黄河,2007,29(02):64-65.
[27]赵志坚.灰色聚类法在水环境质量评价中的应用[J].四川环境,1997,16(03):49-51.
[28]贺青.水质评价的宽域灰色决策法[J].云南环境科学,1995.3,14(1).
[29]李如忠.水质综合评价灰色关联模型的建立与应用[J].安徽建筑工业学院学报,2002,10(1)
[30]郭劲松,王红,龙腾锐.水资源水质评价方法分析与进展[J].重庆环境科学,1999,21(06):6-9.
[31]陈武,艾俊哲.地下水水质综合评价方法探讨[J].地下水, 2002.24(2):74-75.
[32]汪家权,刘万茹,钱家忠,孙世群.基于单因子污染指数地下水质量评价灰色模型[J].合肥工业大学学报(自然科学版),2002,25(05):697-702.
[33]Bin Zhang et al.Prediction of Water Runoff Using Bayesian Concepts and Modular Neural Network[J].Water Resources Research.2000.Vol.36(3):753-761.
[34]薛建军,姚桂基.人工神经网络在水质评价中的应用[J].水文.1997.(3):37-39.
[35]王李管,贾明涛.水质评价的神经网络方法[J].环境工程.1998.16(2):62-65.
[36]胡明星,郭玲香,郭达志.湖泊水质富营养化评价的多准则神经网络法[J].上海环境科学.1998.17(4):14-16.
[37]黄辉金.概率法评价邕江河段水质的尝试[J].广西水利水电,2001,(02):65-67.
[38]陈永灿,郑敬云,刘昭伟.三峡坝区河段水质评价与分析[J].水利水电技术,2001,32(7):24-27.
[39]刘会娟,曲久辉.三峡库区城市江段总体水环境质量综合评价[J].环境科学,2002,23(01):74-77.
[40]戴润泉,臧小平,邱光胜.三峡水库蓄水前库区水质状况研究[J].长江流域资源与环境,2004,13(02):124-127.
[41]罗专溪,张远,郑丙辉,富国,陆兆华.三峡水库蓄水初期水生态环境特征分析[J].长江流域资源与环境,2005,14(06):781-785.
[42]曹明,蔡庆华等.三峡水库库首初期蓄水前后理化因子的比较研究[J].水生生物学报,2006,30(01):12-19.
[43]张继顺,赵南山,毛小波.三峡水库135m蓄水过程分析[J].水力发电,2003(12):19-21.
[44]熊中福.三峡工程蓄水对库区水质的影响[D].重庆:西南大学环境工程,2007.
[45]李公哲.水质管理之原理[M].台北:大学图书出版社,1984.
[46]陈永灿,陈燕,郑敬云,高千红.概率神经网络水质评价模型及其对三峡近坝 水域的评价与水质分析[J].水利发电学报,2004.3:7-15.
[47]饶群.大型水体富营养化数学模拟的研究[D].河海大学水文学及水资源专业,2002.
[48]兰峰.三峡库区富营养化趋势分析[D].重庆:西南师范大学环境科学,2005.
[49]陈永灿,陈燕,郑敬云,高千红.概率神经网络水质评价模型及其对三峡近坝水域的水质评价分析[J].水力发电学报,2004,(03):51-55.
[50]陈明曦,刘德富,黄钰铃.香溪河水华期水质的灰色关联分析[J].三峡大学学报(自然科学版),2006,28(06):552-556.
[51]李柞泳,丁晶,彭荔红著.环境质量评价原理与方法[M].北京:化学工业出版社,2004.
[52]万威涛,水资源质量评价综述[J].水资源研究,2005,12:32-37.
[53]国家环保总局,国家质量监督检验检疫总局.中华人民共和国国家标准《地面水环境质量标准》GB3838-2002.北京中国环境科学出版社,2002:1-10.
[54]飞思科技产品研发中心编著.MATLAB6.5辅助神经网络分析与设计[M].北京:电子工业出版社,2003.
[55]董长虹编著.Matlab神经网络与应用[M].北京:国防工业出版社.2005.1.
[56]苑希民,李鸿雁,刘树坤,崔广涛著.神经网络和遗传算法在水科学领域的应用[M].北京:中国水利水电出版社,2002.
[57]高隽编著.人工神经网络原理及仿真实例[M].北京:机械工业出版社,2003.7.
[58]杨志英.BP神经网络在水质评价中的应用.中国农村水利水电[J],2001.(9):27-29.
[59]黄胜伟,董曼玲.自适应变步长BP神经网络在水质评价中的应用[J].水利学报,2002.(10):119-123.
[60]马细霞,贺晓菊,赵道全等.B-P网络隐含层对水质评价结果的影响分析[J].水电能源科学,2002.20(3):16-18.
[61]黄志洪,武鹏林.基于BP神经网络模型的水质评价方法探讨[J].太原理工大学学报,2005,32(2):174-177.
[62]楼文高,王延政.基于BPR网络的水质综合评价模型及其应用[J],环境污染治 理技术与设备,2003,4(8):23-26.
[63]任黎,董增川,李少华.人工神经网络模型在太湖富营养化评价中的应用[J].河海大学学报(自然科学版),2004,32(2).
[64]李峰.基于BP神经网络的环境质量评价方法[J].中国环境监测,2000,16(3):41-43.
[65]Stambuk-GiljanovicN,Waer Quality Evaluation,by index in Dalmatia[J].Water Research,1999,33(16):3423-3440.
[66]Sharad Kumar Jain.Development of integrated sediment rating curves using ANNs[J].Journal of hydraulic engineering,2001,127(3):181-193.
[67]何镇邦,李桂荣.概率论与数理统计[M].北京:北京理工大学出版社,1988.
[68]谢平,李德,陈广才.基于贝叶斯公式的湖泊富营养化随机评价方法及其验证[J].长江流域资源与环境,2005,3(4).
[69]廖杰,王文圣,丁晶,基于Bayes方法的四川主要河流水质综合评价[J].水文,2007,27(1).
[70]中华人民共和国水利部.中华人民共和国行业标准《水环境监测规范》SL219-98.标准出版社,1998.
[2]国家环境保护总局,2000年中国环境状况公报[J].2000.
[3]高千红.概率神经网络在三峡近坝水域水质评价中的应用研究[D].南京:河海大学,2006.
[4]马倩如,程声通等著.环境质量评价[M].北京:中国环境科学出版社,1990.
[5]环境质量评价原理与方法[M].北京:化学工业出版社,2004.
[6]Horton,R.K.,"An Index-Number System for Rating Water Quality,"Journal of Water Pollution Control Federation,37(3),300-305(1965).
[7]Mance,Elisabeth,An Inexact Two-stage Water Quality Management Model for the He River Basin in Yongxin County,MASc thesis,Jun.2007.
[8]Brown,R.M.,McClelland,N.I.,Deininger,R.A.,Tozer,R.G.,"A Water Quality Index- Do We Dare? ",Water Sewage Works.,117,3139-343(1970).
[9]Harkins,R.S.,"An objective water quality index,"Journal of Water Pollution Control Federation,46(3),588-591(1974).
[10]S.L.Ross.An index system for classifying river water quality.Water Pollut Control(1977),113-122.
[11]张炳文,韩晓滨,杨茂,闫琳.模糊综合指数法在呼和浩特市地面水环境质量综合评价中的应用[J].干旱区资源与环境,1996,10(03):87-92..
[12]李伯昌,施慧燕.长江口河段水环境现状分析[J].水资源保护,2005,21(01):39-44.
[13]沈新强,晁敏.长江口及邻近渔业水域生态环境质量综合评价[J].农业环境科学学报,2005,24(02):270-273.
[14]李柞泳,郭丽婷,欧阳洁.水环境质量评价的普适指数公式[J].环境科学研究,2001,14(3):56-58.
[15]刘征,刘洋.水污染指数评价方法与应用分析[J].南水北调与水利科技,2005.8,3(4).
[16]史宝忠.水质评价的一种新方法一水质综合特征模式[J].中国环境科 学,1994,14(4).
[17]翟由涛,赵玉军.模糊综合指数在判断水质变化趋势和水体管理中的应用[J].环境工程,1995.12,13(6).
[18]陈运东.模糊综合评判法在水质评价中的应用[J].四川水利,1999,20(02):56-58.
[19]徐祖信.我国河流单因子水质标识指数评价方法研究[J]同济大学学报(自然科学版),2005,33(03):321-325.
[20]马成有,曹剑锋,姜纪沂,平建华.改进的尼梅罗污染指数法及其应用-以磐石市地下水环境质量评价为例[J].水资源保护,2006,22(04):53-55.
[21]梁德华.河流水质综合评价方法的统一和改进[J].中国环境监测,2002.18(2):63-66.
[22]邹海明,李粉茹.基于实例的水质模糊综合评价[J].安徽农学通报,2005.11(4):154-155.
[23]郭劲松,龙腾锐等.四种水质综合评价方法的比较[J].重庆建筑大学学报,2000.22(4):6-12.
[24]韩国新,刘明华等.关于区域环境质量的复合模糊距离评价方法及应用探讨[J].中国环境监测,1996,12(6):35-39.
[25]梅学彬,王福纲.模糊综合评价法在水质评价中应用与探讨[J].世界地质,2000.19(2):172-177.
[26]曾永,樊引琴等.水质模糊综合评价法与单因子指数评价法比较[J].人民黄河,2007,29(02):64-65.
[27]赵志坚.灰色聚类法在水环境质量评价中的应用[J].四川环境,1997,16(03):49-51.
[28]贺青.水质评价的宽域灰色决策法[J].云南环境科学,1995.3,14(1).
[29]李如忠.水质综合评价灰色关联模型的建立与应用[J].安徽建筑工业学院学报,2002,10(1)
[30]郭劲松,王红,龙腾锐.水资源水质评价方法分析与进展[J].重庆环境科学,1999,21(06):6-9.
[31]陈武,艾俊哲.地下水水质综合评价方法探讨[J].地下水, 2002.24(2):74-75.
[32]汪家权,刘万茹,钱家忠,孙世群.基于单因子污染指数地下水质量评价灰色模型[J].合肥工业大学学报(自然科学版),2002,25(05):697-702.
[33]Bin Zhang et al.Prediction of Water Runoff Using Bayesian Concepts and Modular Neural Network[J].Water Resources Research.2000.Vol.36(3):753-761.
[34]薛建军,姚桂基.人工神经网络在水质评价中的应用[J].水文.1997.(3):37-39.
[35]王李管,贾明涛.水质评价的神经网络方法[J].环境工程.1998.16(2):62-65.
[36]胡明星,郭玲香,郭达志.湖泊水质富营养化评价的多准则神经网络法[J].上海环境科学.1998.17(4):14-16.
[37]黄辉金.概率法评价邕江河段水质的尝试[J].广西水利水电,2001,(02):65-67.
[38]陈永灿,郑敬云,刘昭伟.三峡坝区河段水质评价与分析[J].水利水电技术,2001,32(7):24-27.
[39]刘会娟,曲久辉.三峡库区城市江段总体水环境质量综合评价[J].环境科学,2002,23(01):74-77.
[40]戴润泉,臧小平,邱光胜.三峡水库蓄水前库区水质状况研究[J].长江流域资源与环境,2004,13(02):124-127.
[41]罗专溪,张远,郑丙辉,富国,陆兆华.三峡水库蓄水初期水生态环境特征分析[J].长江流域资源与环境,2005,14(06):781-785.
[42]曹明,蔡庆华等.三峡水库库首初期蓄水前后理化因子的比较研究[J].水生生物学报,2006,30(01):12-19.
[43]张继顺,赵南山,毛小波.三峡水库135m蓄水过程分析[J].水力发电,2003(12):19-21.
[44]熊中福.三峡工程蓄水对库区水质的影响[D].重庆:西南大学环境工程,2007.
[45]李公哲.水质管理之原理[M].台北:大学图书出版社,1984.
[46]陈永灿,陈燕,郑敬云,高千红.概率神经网络水质评价模型及其对三峡近坝 水域的评价与水质分析[J].水利发电学报,2004.3:7-15.
[47]饶群.大型水体富营养化数学模拟的研究[D].河海大学水文学及水资源专业,2002.
[48]兰峰.三峡库区富营养化趋势分析[D].重庆:西南师范大学环境科学,2005.
[49]陈永灿,陈燕,郑敬云,高千红.概率神经网络水质评价模型及其对三峡近坝水域的水质评价分析[J].水力发电学报,2004,(03):51-55.
[50]陈明曦,刘德富,黄钰铃.香溪河水华期水质的灰色关联分析[J].三峡大学学报(自然科学版),2006,28(06):552-556.
[51]李柞泳,丁晶,彭荔红著.环境质量评价原理与方法[M].北京:化学工业出版社,2004.
[52]万威涛,水资源质量评价综述[J].水资源研究,2005,12:32-37.
[53]国家环保总局,国家质量监督检验检疫总局.中华人民共和国国家标准《地面水环境质量标准》GB3838-2002.北京中国环境科学出版社,2002:1-10.
[54]飞思科技产品研发中心编著.MATLAB6.5辅助神经网络分析与设计[M].北京:电子工业出版社,2003.
[55]董长虹编著.Matlab神经网络与应用[M].北京:国防工业出版社.2005.1.
[56]苑希民,李鸿雁,刘树坤,崔广涛著.神经网络和遗传算法在水科学领域的应用[M].北京:中国水利水电出版社,2002.
[57]高隽编著.人工神经网络原理及仿真实例[M].北京:机械工业出版社,2003.7.
[58]杨志英.BP神经网络在水质评价中的应用.中国农村水利水电[J],2001.(9):27-29.
[59]黄胜伟,董曼玲.自适应变步长BP神经网络在水质评价中的应用[J].水利学报,2002.(10):119-123.
[60]马细霞,贺晓菊,赵道全等.B-P网络隐含层对水质评价结果的影响分析[J].水电能源科学,2002.20(3):16-18.
[61]黄志洪,武鹏林.基于BP神经网络模型的水质评价方法探讨[J].太原理工大学学报,2005,32(2):174-177.
[62]楼文高,王延政.基于BPR网络的水质综合评价模型及其应用[J],环境污染治 理技术与设备,2003,4(8):23-26.
[63]任黎,董增川,李少华.人工神经网络模型在太湖富营养化评价中的应用[J].河海大学学报(自然科学版),2004,32(2).
[64]李峰.基于BP神经网络的环境质量评价方法[J].中国环境监测,2000,16(3):41-43.
[65]Stambuk-GiljanovicN,Waer Quality Evaluation,by index in Dalmatia[J].Water Research,1999,33(16):3423-3440.
[66]Sharad Kumar Jain.Development of integrated sediment rating curves using ANNs[J].Journal of hydraulic engineering,2001,127(3):181-193.
[67]何镇邦,李桂荣.概率论与数理统计[M].北京:北京理工大学出版社,1988.
[68]谢平,李德,陈广才.基于贝叶斯公式的湖泊富营养化随机评价方法及其验证[J].长江流域资源与环境,2005,3(4).
[69]廖杰,王文圣,丁晶,基于Bayes方法的四川主要河流水质综合评价[J].水文,2007,27(1).
[70]中华人民共和国水利部.中华人民共和国行业标准《水环境监测规范》SL219-98.标准出版社,1998.