摘要
水是万物生命之源,水环境管理的好坏直接影响到人类的生存与发展。而水环境质量评价是水环境管理所有工作的基础,传统的评价方法如单因子评价法和综合污染法,因其应用的局限性而备受质疑。因此,寻求一种客观、通用的水质评价方法显得尤为重要。近年来,BP神经网络在模式识别方面的突出表现为之带来了可能。将BP神经网络应用在水质评价中,可以克服传统评价方法的缺点,为各个河流水质类别的纵向对比提供了可能。但由于BP网络自身的缺陷与水质评价的特殊性,使得BP网络水质评价模型面临的两大问题——工作效率与识别精度问题,尚未得到很好的解决。本文围绕这两大问题进行了探索,对改进BP神经网络在水质评价中的应用进行了深入研究。本文的研究工作主要分为以下几个部分:
(1)介绍了BP神经网络的基本理论知识,针对BP网络存在的三个缺陷以及其在水质评价中所遇到的问题,对现有的黄金分割算法进行了改进,将其用于寻找最优隐含层节点个数,达到了优化网络的目的。接着用LM算法对BP网络进行了改进,建立了基于LM-BP网络的水质评价模型,运用该模型对成都市新都区境内流域水质做出了评价,通过与综合污染指数法的评价结果比较,证明了该网络模型的可行性。
(2)为了进一步增强网络的识别精度,将遗传算法与BP网络结合,运用遗传算法的全局寻优能力为BP网络寻找最优的权值与阈值,以此建立GA-BP网络水质评价模型。实验证明,该模型的网络性能(收敛速度与测试样本均方误差)均优于LM-BP网络模型。最后,利用GA-BP网络模型对同一实例进行检测,并分别与LM-BP网络评价模型及综合污染法的评价结果进行对比,证明了GA-BP网络水质评价模型更合理、实用。
(3)为了探讨水质指标与类别之间蕴含的特殊关系,用线性插值取代随机插值生成样本,对已建立的GA-BP网络进行训练,通过实例检测结果对比,表明线性插值的评价结果不能反映水体污染的基本情况。由此证明了用随机插值生成训练样本最能体现水质指标与类别之间复杂的非线性关系。
(4)以上研究表明,本文所建立的基于随机插值生成样本的GA-BP网络水质评价模型识别精度高,实用性和通用性强。最后,通过MATLAB R2009a实现了基于改进BP网络的水质评价模型人机交互界面(GUI),使BP网络水质评价模型从理论研究向实际应用又迈进了一步。
Water is the source of life, and water environmental management has direct impact on humanity’s survival and development. Water environmental quality assessment is the basis of water environmental management. The traditional evaluation methods, such as single-factor evaluation and comprehensive pollution evaluation, are questioned because of their application limitations. Therefore, it’s very important for us to find an objective and universal water quality evaluation method. In recent years, the outstanding performance of BP neural network in pattern recognition makes it possible. The BP neural network used in water quality evaluation can overcome the shortcomings of traditional evaluation methods, and makes it possible for all kinds of rivers to compare water quality longitudinally. Because of the BP network’s defects and the particularity of water quality assessment, the problems of work efficiency and recognition accuracy have not well resolved to the water quality assessment model based on BP network. To solve these problems, this paper researched into water quality assessment model based on improved BP neural network. Main works are as follows:
(1)The basic theory of BP neural network was introduced. Taking the BP network’s defects and its problems met in water quality assessment into account, the Golden Section Algorithm was improved to get the reasonable number of BP neural network's hidden nodes. Then the BP network was improved by the LM algorithm, and a water quality assessment model based on LM-BP network was established. The optimal model was used to evaluate the water quality degree of Xindu area of Chengdu, which were compared with the results derived from comprehensive pollution evaluation method. The feasibility of the water quality assessment model, founded by LM-BP neural network, was proved.
(2)In order to further improve the recognition accuracy of the network, the Genetic Algorithm and BP network were combined. By using genetic algorithm’s global search capability to find the optimal weights and threshold for the BP network, and the Water Quality Assessment based on GA-BP Network was established. Experiments indicate that the model’s network performance (convergence speed and the mean square error of test samples) are better than the LM-BP network model’s. Finally, the GA-BP network model was used to detect the same data as above, and the evaluation results were compared separately with the results from the LM-BP network model and the comprehensive pollution assessment method. The comparison showed that the GA-BP network water quality assessment model is more reasonable and practical than the model based on the LM-BP network.
(3)In order to clarify the special relationship between the water quality indexes and water quality grades, linear interpolation was used to generate enough samples instead of random interpolation to train the founded GA-BP network. Comparing the test results, it was proved that the water evaluation grades can’t reflect the basic pollution condition of the river. This indicated that the training samples generated by random interpolation well revealed the complex nonlinear relationship between the water quality indexes and the water quality grades.
(4)All studies above showed that the GA-BP network water quality assessment model trained by samples generated by random interpolation has high recognition accuracy, practicality and versatility. Finally, the Graphical User Interface (GUI) of the water quality assessment model based on improved BP network was set up by MATLAB R2009a, which accelerated the progress of the BP network water quality assessment model from the theory to the practical application.
引文
[1]吴红燕.水质监测在水资源保护中的作用[J].电力学报,2007,22(4):509-511.
[2]谈鹏程.基于规则的知识系统在水质评价中的应用研究[D].河海大学,2006.
[3]彭文启,张祥伟,等.现代水环境质量评价理论与方法[M].北京:化学工业出版社,2005.
[4]张尧旺.水质监测与评价[M].河南:黄河水利出版社,2005.
[5]张锁春.环境科学中的数学理论与方法[J].环境科学丛刊,1992,13(2) :12-17.
[6]劳期团.模糊数学方法在水库水质综合判别中的应用[J].中国环境科学,1989,9(3):225-228.
[7]慕金波,等.灰色聚类法在水环境质量评价中的应用[J].环境科学,1981, 12(2): 86-90.
[8]邓聚龙.灰色预测与决策[M].华中理工大学出版社,1988.
[9]郭小青,项新建,等.基于神经网络模型的水质监测与评价系统[J].重庆环境科学,2003,25(5):8-10.
[10]王晓玲,李松敏,等.基于遗传神经网络模型的水质综合评价[J].中国给水排水,2006,22(11):45-48.
[11]陈守煜,李亚伟.基于模糊人工神经网络识别的水质评价模型[J].水科学进展,2005,16(1):88-90.
[12]杨芳,原松.基于BP神经网络的水环境质量评价模型的研建[J].人民长江,2008,39(23):46-48.
[13]秦传玉,赵勇胜,等.基于BP神经网络的齐齐哈尔地区地下水水质评价[J].环境监测管理与技术,2007,19(2):15-18.
[14]韩波,孙利,等.水质评价模式识别的BP神经网络方法[J].广州环境科学,2005,20(4):40-43.
[15]李峻,孙世群.BP神经网络在青戈江水质评价上的应用[J].安徽建筑,2008,160(3):167-172.
[16]娄申,干晓蓉.基于BP神经网络的水质评价[J].云南名族大学学报,2007,16(2):165-167.
[17]高娟,华珞,等.地表水水质监测现状分析与对策[J].首都师范大学学报(自然科学版),2006,27(1):75-80.
[18]马倩如,程声通,等.环境质量评价[M].北京:中国环境科学出版社,1990.
[19]樊亚文,马娟.基于黑板结构的信息融合专家系统在水质监测中的应用[J].安徽水利水电职业技术学院学报,2006,6(2):7-9.
[20]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.
[21]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.
[22]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.
[23]陈武阳,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.
[24]朱静平.几种水环境质量综合评价方法的探[J].西南科技大学学报,2002,17(4):62-67.
[25]薛建军,姚桂基,等.人工神经网络在涅水水质综合评价中的应用[J].青海环境,1997,7(1):25-28.
[26]任黎,董增川.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,2004,32(2):148-150.
[27]郭劲松,土红,等.水环境质量的隶属度BP网络决策模型[J].中国给水排水,2000,16(3):9-11.
[28] T. R. Neelakantan, N. V. Fundarikanthan. Neural network based simulation optimization model for reservoir operation [J]. Journal of water resources planning and management, 2000, 26(2): 57-64.
[29] Marina Campolo et a1. Forecasting river flow rate during low flow period using neural network [J]. Water resources research, 1999, 35(I 1): 3547-3552.
[30]李继选,王军.水环境数学模型研究进展[J].水资源保护,2006,22(1):9-13.
[31]金明.一维稳态河流的随机微分方程模型[J].水利学报,1991(2):19-25.
[32] Srinivas M, Patnaik L M. Adaptive probability of crossover and mutation in genetic algorithm. IEEE Trans on SMC, 1994, 24(4): 656-667.
[33]李海英,秦肖生.综合性遗传算法用于水质模型参数估值[J].中国给水排水,2002,18(5):28-30.
[34]申玮,郭宗楼,等.模拟退火算法在水质模型参数识别中的应用[J].西北农林科技大学学报,2004,32(5):101-105.
[35] Sharad KJ. Development of integrated sediment rating curves using ANNS [J]. Journal of Hydraulic Engineering, 2001, 127(1): 30-37.
[36] Marina C. orecasting fiver flow rate during low flow period using neural network [J]. Water Resources Research, 1999, 35(11): 3547-3552.
[37]魏文秋,孙春鹏.灰色神经网络水质预测模型[J].武汉水利电力大学学报,1998,31(4):26-29.
[38] ZHANGB. Prediction of water run off using Bayesian concepts and modular neural network [J]. Water Resources Research, 2000, 36(3): 753-762.
[39] Fai, I, C. Generalized model of pentachlorophenol distribution in a mended soil water system [J]. Water Environment Research, 2001, 73(1): 110-117.
[40] WASP4: A hydrodynamic and water quality model. EPA, 1988.
[41] Shanahan E. River water quality model no.1 (RWQMI): In modeling approach. Water Science Tech, 2001, 43(5): 1-10.
[42]谭现虎.基于多Agent的水质监测系统研究与实现[D].中国科学技术大学,2006.
[43] Karul C, Soyupak S, Ahmet FC. Case studies on the use of neural networks in eutrophication modeling [J]. Ecological Modeling, 2000, 134: 145-152.
[44]邹永祥.基于神经网络的信息融合技术与应用[D].成都理工大学,2008.
[45]李晶,张征,朱建刚.BP神经网络地下水水质评价若干问题及修正[J].环境科学与技术,2009,32(6C):387-392.
[46]陈兴,程吉林,刘芳.BP神经网络用于水质评价的参数确定[J].水利与建筑工程学报,2007,5(1):12-15.
[47]刘娟,蒋兆华,韦军,凌启东.BP神经网络在地下水水质评价中的应用[J].水利科学与经济,2009,15(3):191-193.
[48]蒋佰权,王万森,温香彩.改进算法的BP神经网络水质评价模型[J].计算机系统应用,2007,(9):46-50.
[49]张文范,张伟.基于改进BP人工神经网络在地下水水质现状评价中的应用[J].吉林水利,2008,9(316):31-33.
[50]宋国浩.人工神经网络在水质模拟与水质评价应用中的研究[D].重庆大学,2008.
[51]邹红娟,林子扬,等.人工神经网络方法在资源与环境预测方面的应用[J].长江流域资源与环境,2000,19(2):29-31.
[52]刘进涛.模糊神经网络在水质评价中的应用[D].首都师范大学,2006.
[53]李敏生,刘斌.BP学习算法的改进与应用[J].北京理工大学学报,1999,19(6):721-724.
[54] R. A. Jacobs. Increased rates of convergence through learning rate adaptation [J]. Neural Networks, 1988, 1(4): 295-307.
[55] S. C. Ng, C. C. Cheung, S. H. Leung.Magnified Gradient Function with Deterministic Weight Modification in Adaptive Learning [J].IEEE Trans on Neural Networks, 2004, 15(6): 1411-1423.
[56]Y.F. Yam, W S. Chow. A weight initialization method for improving training speed in feed-forward neural network [J]. Neurocomputing, 2000, 30: 219-232.
[57] Y. F. Yam, C. T. Leung, P. K. S. Tam. An independent component analysis based weightinitialization method for multiplayer perceptrons [J]. Neurocomputing, 2002, 48: 807-818.
[58] Zhihong Yao, Minrui Fei, Kang Li. Recognition of blue-green algae in lakes usingdistributive genetic algorithm-based neural networks[J]. Neurocomputing, 2007, 70: 641-647.
[59]左志强,陈永进,慈春令.基于MATLAB下的改进BP算法[J].燕山大学学报,2001,25(12):101-104.
[60]张彤,肖南峰.基于BP网络的数字识别方法[J].重庆理工大学学报(自然科学版) 2010(3):47-51.
[61]张彤,卢雯雯,肖南峰.基于BP网络的人脸朝向识别方法[J].重庆理工大学学报(自然科学版) 2010(6):65-69.
[62] G D. Magoulas, M. N.Vrahatis, G. S. Androulakis. Effective back-propagation with variable step size [J]. Neural Networks, 1997, 10(1): 69-82.
[63]韩力群.人工神经网络理论设计及应用[M].北京:化学工业出版社,2002.
[64]张海燕,冯天瑾.新的组合激活函数BP网络模型研究[J].青岛海洋大学学报,2002,32(4):621-626.
[65]潘涛.BP神经网络模型中活化函数对网络性能影响的研究[J].安徽师范大学报(自然科学版),1998,21(3):218-221.
[66]张彤,肖南峰.基于BP网络的指纹识别系统[J].重庆理工大学学报(自然科学版) 2010(1):51-54
[67]闫文秀,裴建岗,孙颖.基于Gabor滤波器和改进BP神经网络的人脸检测方法[J].重庆工学院学报(自然科学版) 2009(4):104-108.
[68]杨安华,彭清娥,等.BP算法固定学习率不收敛原因分析及对策[J].系统工程理论与实践,2002,22(12):22-25.
[69]魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版社,2005.
[70]徐冠,夏克文,等.基于神经网络的冠心病智能诊断系统[J].计算机工程与应用,2005,35(03):197-199.
[71]黄勤,常伟,刘益良,刘衍鹏.基于PCA的BP神经网络分类器[J].重庆工学院学报(自然科学版) 2009(7):95-102.
[72]高富强,邹恒,秦昌硕.BP和RBF神经网络在字母识别中的比较[J].重庆工学院学报(自然科学版) 2009(9):83-86.
[73]杨芳,原松.基于BP神经网络的水环境质量评价模型的研建[J].人民长江,2008,39(23):46-48.
[74]李晶,张征,朱建刚.BP神经网络地下水水质评价若干问题及修正[J].环境科学与技术,2009,32(6C):387-392.
[75]夏克文,李昌彪,等.前向神经网络隐含层节点数的一种优化算法[J].计算机科学,2005,32(10):143-145.
[76]张华节,李璀.学习速率连续变化的BP神经网络算法[J].四川兵工学报,2011(4):61-64.
[77]张方方,张莲,刘贺.基于神经网络的室内环境热舒适度融合算法[J].重庆工学院学报(自然科学版) 2009(10):120-124.
[78]李晔.基于排序选择的改进遗传算法优化的神经网络控制[J].四川兵工学报,2010(9):117-119.