類神經網路於環境資訊之鑑識、推估及預測之研究
論文摘要
本研究探討類神經網路在環境資訊上之應用。包括如何鑑識環境資訊、如何推估環境資訊與如何預測環境資訊。研究內容可概分成四項，第一為水庫水質優養資訊之鑑識，研究案例是以負責供應大台北地區生活用水之翡翠水庫為對象，以模糊理論與類神經網路分析水庫水質優養情況，並與Carlson優養指標交互驗證，研究結果顯示模糊綜合評估法在1989、1991及1992年的10月都評判出翻轉效應所造成的優養化現象，但Carlson法卻無法判別出。而自組織映射網路(Self-Organizing Feature Map, SOFM)也同樣地鑑識到翻轉現象之發生，且SOFM網路輸出圖上可明顯分出三個群集，在對角線上靠近最左下角的位置是屬於貧養狀態辨識區，對角線的中央部分是普養狀態辨識區；對角線最右上方的區域則為優養辨識區。此外，SOFM輸出圖上還可發現異常數據辨識區，而落於此區之數據皆是Carlson法與模糊綜合評判法結果相異之數據點。針對異常數據點，本研究推論可能與聖嬰現象具有微相關之關連性。
其次為土壤環境中污染物分佈資訊之推估，研究一個虛擬受污染之土壤場址，根據少量之採樣點，以類神經推估法將整個受污染場址之濃度分佈情形計算出，並與一般克利金法比較推估精準度，研究結果發現以類神經推估法計算未採樣點之濃度值，必須將網路訓練至方均根誤差小於10-6。若於訓練過程中，方均根誤差無法收斂至10-4以下，則該採樣資料集屬於不具空間相關性之結構，因此將無法進行空間推估，且類神經推估法用20個採樣點即得到良好之推估結果，一般克利金法則否。總結而論，類神經推估法之推估正確率高於一般克利金法，且發生正錯推估及負錯推估之百分率小於一般克利金法。
第三為模場中降雨逕流內可溶性物質濃度資訊的預測，研究對象是實驗室中的逕流實驗，先以類神經路中之倒傳遞網路建一預測模式，此模式可預測模場逕流出流水中可溶性物質之濃度，然於初始濃度低之數組實驗，該模式之預測結果有低估之現象，因此本研究提出一種能提高類神經網路預測準確率的理論-動態建模理論，研究結果指出依動態建模理論所建之模式較單一類神經網路之模式更能預測此系統之行為。最後藉數學推導之方式解釋模糊系統與類神經網路是等價關係，藉此說明類神經網路累積知識之方式與結果，不是一個黑箱過程，並以水庫水質優養之評價為例，闡述從訓練完成後之類神經網路中至少可擷取出十條以上可用作水質評鑑之模糊規則。
關鍵詞：類神經網路、模糊理論、環境資訊、優養化、地理統計、降雨逕流。

Characterization, Estimation, and Prediction of Environmental Information using Artificial Neural Networks
ABSTRACT
The research was based on the techniques of artificial neural networks (ANN) to explore the feasibility of applications of ANN techniques in recognizing, estimating, and predicting environmental information. This thesis was conducted in four parts. The first part of the study explored the characterization of reservoir water quality information using fuzzy theory and ANN technique. A general methodology for fuzzy synthetic evaluation and a self-organizing feature map (SOFM) model are developed and illustrated with a case study of trophic status assessment for Fei-Tsui Reservoir which supply for the civil water demands in great Taipei area. The results showed that the long-term change of water quality and the overturn phenomena cannot be observed with the Carlson index from 1987 to 1992 but is expressed by fuzzy synthetic evaluation and SOFM model. On the output map of SOFM model, 3 clusters can be observed. The cluster occupied at the lower-left corner of the map expressed those mapping data as oligotrophic state; the cluster occupied at center of the axis of the map indicated those mapping data as mesotrophic state; the cluster occupied at the upper-right corner of the map expressed those data as eutrohpic state.
The second part of the study is about the estimation of the pollutant concentration in a virtual contaminated soil site. A neural estimation model is described, implemented in a parallelizing algorithm, and applied to develop maps of discrete spatially distributed fields. A comparison with a reference field shows that neural estimation produces more correct estimation than does ordinary kriging.
The third part of the study explored the prediction of the solute concentration in surface runoff. The runoff experiments were performed in a laboratory flume with a medium-packed bed. The relationships between the controlled input variables and the affected outflow concentration were modeled using ANN models and a diffusion model. The results showed that a single BPN model can mostly predict the outflow concentration well except those experiments with lower initial concentration in medium-packed bed. For improving the predicting accuracy, dynamic modeling theory (DMT) was proposed in this study. The results revealed that a model with multiple neural networks created by DMT is more suitable than a single BPN model for predicting the outflow concentration of the rainfall-runoff experiments. In the last section of the study, we provided a mathematical interpretation to express the equivalence relationship between ANN and a fuzzy system. Using the interpretation, ANN would no longer be seen as a black-box theory. In addition, Fei-Tsui Reservoir was used as an example to extract 10 fuzzy rules from a trained neural network for testing the knowledge-extracting procedure.
Keywords: Artificial Neural Networks, Environmental Information, Eutrophication, Geostatistics, rainfall-runoff

封面
目錄
第一章 緒論
1-1 研究動機與目的
1-2 研究內容
1-3 本文結構
第二章 文獻回顧
2-1 類神經網路
2-2 類神經網路應用相關文獻分析
2-2-1 ANN在空氣方面之應用
2-2-2 ANN在癈水方面之應用
2-2-3 ANN在噪音部份之應用
2-2-4 ANN在土壤方面之應用
2-2-5 ANN在淨水工程方面之應用
2-2-6 ANN在吸附、分離上之應用
2-2-7 ANN在自然資源方面之應用
2-3 環境資訊之相關研究
2-3-1 水環境中水資鑑之相關研究
2-3-2 土壤環境中污染物濃度分佈資訊之推估
2-3-3 逕流環境中水質資訊預測之相關研究
第三章 類神經網路與模糊理論於水質資訊之鑑識
3-1 模糊理論
3-1-1 定義模糊集合
3-1-2 尋求隸屬函數
3-1-3 權重之決定
3-1-4 糢糊綜合評價
3-1-5 模糊優養指數
3-2 自組織特映射理論(SOFM)
3-2-1 競爭式學習法則
3-2-2 自組織映射網路之學習演算法
3-2-3 自組織映射網路之簡例說明
3-3 以Carlson法進行水質評判
3-4 以模糊綜合評判進行水質評判
3-5 以自組織映射網路進行水質評判
3-6 Carlson法、模糊綜合評判法與SOFM之綜合比較
3-6-1 貧養別區之發現
3-6-2 優養別區的發現
3-6-3 普養別區的發現
3-6-4 異常數據區之發現
3-6-5 SOFM辨識異常數據結果之另一種解釋
3-7 小結
第四章 類神經網路與克利金模式於金模式於土壤環境資訊之推估
4-1 旋轉條帶法TBN之模擬
4-2 克利金模式
4-3 反傳遞網路
4-3-1 網路架構
4-3-2 CPN演算法
4-4 虛擬場址之空間結構分析
4-5 虛擬場址之應用
4-6 類神經推估法與一般克利金之比較
4-7 不同採樣方式類神經推估法之影響
4-8 指標克利金、自然對數轉換法與神經推估法之比較
4-9 一般克利金與類神經推估法之風險分析
4-10 小結
第五章 類神經網路與定率模式於逕流環境資訊之預測
5-1 玻璃珠介質逕流實驗
5-2 擴散模式之理論
5-2-1 模式之基本假設
5-2-2 模式之建立
5-3 類神經網路與動態建模理論
5-4 三種模式應用於模擬真實土壤之溶質流失
5-4-1 真實土壤逕流實驗
5-4-2 以沖刷模式模擬土壤之溶質流失
5-4-3 以擴散模式模擬土壤之溶質流失
5-4-4 類神經網路預測土壤之溶質流失
5-5 小結
第六章 類神經網路與模糊理論之價關係
6-1 類神經網路與模糊加成系統之等價證明
6-2 糢糊規則擷取
6-2-1 糢糊規則的描述
6-2-2 類神經糢糊規則之轉譯
6-3 從類神經糢糊中擷取水庫質評判規則
第七章 結論與建議
7-1 結論
7-2 建議
附錄A 聖嬰現簡介
附錄B Carlson指標優養評表
附錄C 模糊綜合評判表
附錄D 虛擬場址之全域資料

參考文獻
Abdulbary, A. F.; Lai, L. L.; Al-Gobaisi, D. M.K.; Husain, A. (1993). Experience of using the neural network approach for identification of MSF desalination plants. Desalination, 92(1-3), 323-332.
Ahuja, L. R. and Lehman, O. R. (1983). The extent and nature of rainfall-soil interaction in the release of soluble chemicals to runoff. J. Environ. Qual., 12, 34-40.
Ambrose, R. B., Barnwell, T. O. (1989). Environmental Software at the U. S. Environmental Protection Agency*s Center for Exposure Assessment Modeling,, Environment Software, 4(2), 76-93.
Basheer, I A; Reddi, L N; Najjar, Y M. (1996). Site Characterization by Neuronets: An Application to the Landfill Siting Problem. Ground water, 34(4), 610-619.
Basheer, I. A.; Najjar, Y. M.; Hajmeer, M. N. (1996). Neuronet Modeling of VOC Adsorption by GAC. Environmental technology, 17(8), 795-804.
Basheer, I. and Najjar, Y. M. (1996). Predicting dynamic response of adsorption columns with neural nets. J. Comp. in Civil Engrg, ASCE, 10(1), 31-39.
Basheer, I. and Najjar, Y. M., and Swartz, S. (1994). Predicting permeability of compacted clay using an artificial neural network. Intelligent engineering systems through artificial neural networks. 4, 1161-1166.
Benitez, J. M., Castro, J. L. and Requena, I. (1997). Are Artificial Neural Networks Black Boxes?. IEEE Transactions on Neural Networks, 8(5), 1156-1164.
Bennett, C. O., Myers, J. E. (1974). Momentum, Heat, and Mass Transfer, New York, McGraw-Hill Book Co.
Boger, Zvi. (1992). Application of Neural Networks to Water and Wastewater Treatment Plant Operation. ISA transactions : a publication of Instrument Society of America, 31(1), 25-34.
Cal, Y (1995). Soil classification by neural network. Advances in engineering software, 22(2), 95-98.
Cammarata, G., Cavalieri, S. and Fichera, A. (1995). Neural network architecture for noise prediction, Neural Networks 8(6), 963-973.
Capodaglio, A., Jones, H. (1992). Real Time Control of wastewater treatment operations, Water Science and Technology, 25, 89-101.
Chen, S., Billing, S. A. and Grant, P. M. (1990). Non-linear system identification using neural networks, Int. J. Control, 51(6), 1191-1214.
Chuah, H. T., (1993). Retrieval of soil moisture content from radar backscatter coefficients, International Geoscience and Remote Sensing Symposium (IGARSS), 1, 338-340.
Coit, David W; Smith, Alice E. (1996). Solving the redundancy allocation problem using a combined neural network/genetic algorithm approach. Computers & operations research, 23(6), 515-524.
Conlin, J.; Peel, C.; Montague, G. A. (1997). Modelling pressure drop in water treatment. Artificial intelligence in engineering, 11(4), 393-402.
Deutsch, C. V., A. G. Journel, (1992). GSLIB: Geostatistical Software Library ans User*s Guide. Oxford University Press, New York, N. Y.
El-Hawary, M. E. (1993). Artificial neural networks and possible applications to desalination. Desalination, 92(1-3), 125-148.
Enbutsu, I.; Baba, K.; Hara, N.; Waseda, K.; Nogita, S. (1993). Integration of multi AI paradigms for Intelligent Operation Support Systems--fuzzy rule extraction from a neural network. Water science and technology, 28(11-12), 333-342.
Flatman, G. T., Englund, E. J. and Yfantis, A. A. (1988). Geostatistical Approaches to the Design of Sampling Regimes, Princoples of Environmental Sampling, L. H. Keith, ed., American Chemistry Society, Inc., Washington, D. C., 73-84.
Fleming, N. S. (1994). Comparison of water consumption forecasting methods including artificial neural networks. National conference publication - Institution of Engineers, Australia.,15(3), 115-124.
Flood, I. and Kartam, N. (1994). Neural networks in civil engineering. I: Principles and understanding. J. Comp. in Civil Engrg, ASCE, 8(2), 131-148.
French, M. N., Krajewski, W. F., and Cuykendall, R. R. (1992). Rainfall Forecasting in Space and Time Using a Neural Network, J. of Hydrology, 137, 1-31.
Frith, A. (1996). Casting Neural Nets in Water: The applications of artificial neural networks and their use in the water industry. Process engineering, 77(1), 67-73..
Gagnon, C., Grandjean, B. P. A., and Thibault, J., (1997). Modelling of Coagulant Dosage in a Water Treatment Plant, Artificial Intelligence in Engineering, 11, 401-404.
Gagnon, C.; Grandjean, B. P. A.; Thibault, J. (1997). Modelling of coagulant dosage in a water treatment plant. Artificial intelligence in engineering, 11(4), 401-408.
Gao, C; Govind, R; Tabak, H H. (1996). Predicting Soil Sorption Coefficients of Organic Chemicals Using a Neural Network Model. Environmental toxicology and chemistry, 15(7), 1089-1098.
Garrett, J. H., Ranjitham, S. and Eheart, J. W. (1992). Application of neural network to groundwater remedaition; Chapter 11. Expert systems for civil engineers: knowledge representation, R. Allen, ed., 1259-1269.
Goh, Anthony T C; Attoh-Okine, N O. (1997). Modeling Soil Correlations Using Neural Networks. Journal of computing in civil engineering, ASCE, 11(1), 79-88.
Goh, T. C. (1995). Modeling soil correlations using meural networks. J. Comp. in Civil Engrg., ASCE, 9(4), 275-277.
Grubert, J. P. (1995). Prediction of Estuarine Instabilities with Artificial Neural Networks, J. of Computing in Civil Engineering, 9(4), 266-274.
Journel, A. G. (1988). Nonparametric Geostatistics for Risk and Additional Sampling Assessment, Princoples of Environmental Sampling, L. H. Keith, ed., American Chemistry Society, Inc., Washington, D. C., 45-72.
Juang, K. W., Lee, D. Y. (1998). Simple indicator kriging for estimating the probability of incorrectly delineating hazardous areas in a contaminated site. Environ. Sci. Technol., 32, 2487-2493.
Karunanithi, N., Grenney, W. J., Whitley, D. and Bovee, K. (1994). Neural networks for river flow prediction. J. Comp. in Civil Engrg., ASCE, 8(2), 201-220.
Kastens, Terry L; Featherstone, Allen M. (1996). Feedforward Backpropagation Neural Networks in Prediction of Farmer Risk Preferences. American journal of agricultural economics, 78(2), 400-409.
Kenneth, F. R., Bo Ling (1994). Neural networks with multiple-state neurons for nitrogen oxide (NOx) emissions modeling and advisory control, 1994 IEEE International Conference on Neural nerworks, 6, 3834-3839.
Kock, G; Serbedzija, N B, (1996). Simulation of Artificial Neural Networks. Systems analysis, modelling, simulation, 27(1), 15-24.
Kohonen, T. (1990). The self-organizing map, Proc. IEEE, 78, 1414-1480.
Lai, J. L., Lo, K. (1992). Modelling of Solute Transfer to Surface Runoff. Water Science and Technology, 26, 1851-1856.
Lee, Y. -M. (1995). Entropy/Kriging-based formulations for groundwater monitoring network design, Ph.D. dissertation, The Johns Hopkins University, Baltimore, Maryland.
Liao, Hsiu-Hua; Tim, U S. (1997). An Interactive Modeling Environment for Non-Point Source Pollution Control, Water resources bulletin, 33(3), 591-603.
Maier, Holger R; Dandy, Graeme C. (1996). The use of artificial neural networks for the prediction of water quality parameters. Water resources research, 32(4), 1013-1028.
Malkar, P., Bojnar, M. (1994). Short-term air pollution prediction on the basis of artificial neural networks, International Conference on Air Pollution-Proceedings, 1, 545-552.
Marija, B., Lesjak, M. and Mlakar, P. (1993). A neural network-based method for short-term predictions of ambient SO2 concentrations in highly polluted industrial areas of complex terrain, Atmospheric Environment, 27B(2), 221-230.
Mukherjee, Abhijit; Deshpande, Jayant M; Rajjasekaran, S; Ramasamy, J V; Basheer, Imad A; Najjar, Yacoub M. (1997). Modeling Initial Design Process Using Artificial Neural Networks. Journal of computing in civil engineering, 11(2), 145-154.
Nefati, Hedi; Cense, Jean-Michel; Legendre, Jean-Jacques. (1996). Prediction of the Impact Sensitivity by Neural Networks. Journal of chemical information and computer sciences, 36(4), 804-815.
Niemi, H; Bulsari, A; Palosaari, S. (1995). Simulation of membrane separation by neural networks. Journal of membrane science, 102, 185-112.
Pachepsky, Ya A; Timlin, Dennis; Varallyay, G. (1996). Artificial Neural Networks to Estimate Soil Water Retention from Easily Measurable Data. Soil Science Society of America journal, 60(3), 727-737.
Parr. A. D., Richardson, C., Lane, D. D. and Baughman, D. (1987). Pore water uptake by agricultural runoff. J. of Environmental Engineering, ASCE, 113, 49-63.
Potukuchi, S., Wexler, A. (1997). Predicting Vapor Pressures Using Neural Networks, Atmospheric Environment, 31(5), 741-753.
Raman, H.; Sunilkumar, N. (1995). Multivariate modelling of water resources time series using artificial neural networks. Hydrological sciences journal, 40(2), 145-156.
Ranjithan, S., Eheart, J. W. (1993). Neural Network-based Screening for Groundwater Reclamation Under Uncertainty, Water Resources Research, 29(3), 563-574.
Richardson, C. P. and Parr, A. D. (1988). Modified Fickian model for solute uptake by runnoff. J. of Environmental Engineering, ASCE, 114, 792-809.
Roadknight, C. M., Balls, G. R., Mills, G. E. and Dominic Palmer-Brown (1997). Modeling complex environmental data, IEEE Transactions on Neural Networks, 8(4), 852-862.
Rodriguez, M. J.; Serodes, J. B.; Cote, P. A. (1997). Advanced chlorination control in drinking water systems using artificial neural networks. Water supply, 15(2), 159-171.
Rogers, L. L. and Dowla, F. U. (1994). Optimization of groundwater remediation using artificial neural network with solute transfer modeling. Water Resources Research, 30(2), 457-481.
Ruiz-Suarez, J.C., Mayora, O. A., Smith-Perez, R. and Ruiz-Suarez L. G. (1994a). Neural network-based prediction model of ozone for mexico city, International Conference on Air Pollution-proceeding, 1, 393-400.
Ruiz-Suarez, J.C., Mayora, O. A., Smith-Perez, R. and Ruiz-Suarez L. G. (1994b). Short-term forecasting of ozone by means of a bidirectional associative memory, Applications of Artificial Intelligence in Engineering, 33, 107-114.
Ruiz-Suarez, J.C., Mayora, O. A., Smith-Perez, R. and Ruiz-Suarez L. G. (1994b). Short-term ozone forecasting of ozone by artificial neural networks, Advances in Engineering Software, 23, 143-149.
Rydygier, Edward. (1997). Using Neural Networks to Solve Prediction Problems in Econometrics and Economics. Systems analysis, modelling, simulation, 27(4), 289-293..
Schaap, Marcel G; Bouten, Willem. (1996). Modeling water retention curves of sandy soils using neural networks, Water resources research, 32(10), 3033-3048..
Selvaraj, R; Deshpande, P B; Tambe, S S; Kulkarni, B D. (1995). Neural networks for the identification of MSF desalination plants. Desalination, 101(2), 185-192.
Snyder, J. K., Woolhiser, D. A., (1985). Effects of Infiltration on Chemical Transport into Overland Flow, Trans. ASAE, 28, 1450-1457.
Song, X. and Hopke, P. K. (1996). Solving the chemical mass balance problem using an artificial neural network. Environ. Sci. Technol., 30, 531-535.
Spall, J. C., Cristion, J. A. (1994). A neural networks controller for system with unmodeled dynamics with application to wastewater treatment, 1994 IEEE International Symposium on Intelligent Control, 273-278.
Tamari, S; Wosten, J H M; Ruiz-Suarez, J C. (1996). Testing an Artificial Neural Network for Predicting Soil Hydraulic Conductivity. Soil Science Society of America journal, 60(6), 1732-1744.
Teh, C I; Wong, K S; Goh, A T C; Jaritngam, S. (1997). Prediction of Pile Capacity Using Neural Networks. Journal of computing in civil engineering, 11(2), 129-137.
Trejo, L. A., Carlos, S. (1995). Improved back-propagation: Epsilon-back-propagation, From Natural to Artificial Neural Computation, 427-432.
Tsaptsinos, D., Tang, R. and Leigh, J. R. (1995). Neuroidentification of a biotechnological process: Issues and application, Neurocomputing, 9, 63-79.
Vieira, R. S., Nielsen, D. R. and Biggar, J. W., (1981). Spatial Variability of Field-measured Infiltration Rate, Soil Sci. Soc. Am. J., 45, 1040-1048.
Voutchkov, I. I.; Velev, K. D. (1997). Identification of Waste Water Treatment Plant Using Neural Networks. Lecture notes in computer science, 1226, 478-490.
Wallach, R. and Genuchten, M. T. (1990). A physically based model for predicing solute transfer from soil solution to rainfall-induced runoff water. Water Resources Research, 26, 2119-2126.
Wallach, R., Jury, W. A. and Spencer, W. F. (1988). Transfer of chemicals from soil solution to surface runoff : a diffusion-based soil model. Soil Sci. Soc. Am. J., 52, 612-618.
Wienke, D., Hopke, P. K. (1994a). Projection of prim*s minimal spanning tree into a Kohonen neural network for identification of airborne partical source by their multielement trace patterns, Analytical Chemica Acta, 91,1-18.
Wienke, D., Hopke, P. K. (1994b). Uisual neural mapping technique for locating fine airborne particles sources, Environ. Sci. Technol., 28, 1015-1022.
Wienke, D., Hopke, P. K. (1994c). Mutiple site receptor modeling with a minimal spanning tree combined with a neural network. Environ. Sci. Technol., 28, 1023-1030.
Wilcox, S. J., Hawkes, D. L. (1995). A neural network based on bicarbonate monitoring, to control anerobic digestion. Water Research, 29(6), 1465-1470.
Wittmann, C; Schmid, R D; Loffler, S; Zell, A. (1997). Application of a Neural Network for Pattern Recognition of Pesticides in Water Samples by Different Immunochemical Techniques. ACS symposium series, 657, 343-354.
Yang, M; Hubble, J; Lockett, A D; Rathbone, R R. (1996). A Neural Network Model for Prediction of Binary Adsorption Using Single Solute and Limited Binary Solute Adsorption Data. Separation science and technology, 31(9), 1259-1271.
Yeh, S. -C. (1996). Grey programming and its applications to water resources management, Ph.D. dissertation, Cornell University, New York.
Yeh, W. W. G. (1985). Reservoir Management and Operation Models: A State-of- the -Art Review, Water Resiurces Research, 21(12), 1797-1818.
Yost, R. S., Uehara, G., and Fox, R. L. (1982). Geostatistical analysis of soil chemical properties of large land area. II. Kriging. Soil Sci. Soc. Am. J., 46, 1033-1037.
Zadeh, L. A. (1965). Fuzzy sets. Information and Control, 8, 338-353.
Zhang, Q., Stanley, J. S. (1997). Forecasting Raw-water Quality Parameters for The North Saskatchewan River by Neural Network Modeling. Water Resource, 31(9), 2340-2350.
Zurada, J. M. (1992). Introduction to Artificial Neural System. Info Access Distribution Pte LTD.
余瑞芳、陳萬原、廖述良、張鎮南，1996，〝類神經網路於連續流SBR廢水處理系統即時控制之應用〞第二十一屆水處理技術研討會，pp457-464。。
吳信賢，1996，類神經網路在受體模式之應用，國立台灣大學，碩士論文。
胡景堯，1996，模糊理論在水庫優養化判別上的應用，國立台灣大學，碩士論文。
倪榮興，1996，污水處理廠水質管理專家系統之研究，國立雲林技術學院，碩士論文。
徐明麟，1997，克利金法預測土壤重金屬污染範圍，國立台灣大學，碩士論文。
高全興，1997，類神經網路於空氣品質短期預測之研究，國立雲林技術學院，碩士論文。
莊愷偉，1995，以地理統計法預測土壤重金屬含量之研究，國立台灣大學，碩士論文。
黃尚雄，1996，以類神經網路預測空氣品質之研究 國立交通大學，碩士論文。
溫清光、張穗蘋，水庫蓄水初期內部貢獻營養物質對水庫優養化之影響，第八屆環境規劃與管理研討會，台南，36-45。
葉怡成，1994，類神經網路-模式應用與實作儒林圖書有限公司。
盧瑞山、駱尚廉，1995，模糊綜合評判在台灣地區水庫優養化判定上之應用，第八屆環境規劃與管理研討會，台南，66-74。
盧瑞山、駱尚廉，1997，類神經網路應用於溶質傳輸至地表逕流水之研究，第十屆環境規劃與管理研討會論文集，台中，351-357。
賴俊良，1994，溶質進入地表逕流傳輸模式之研究，國立台灣大學，博士論文。
闕頌廉，「應用模糊數學」，科技圖書股份有限公司，1992。
闕蓓德，1997，受重金屬污染土壤復育決策支援系統之雛形研究，國立台灣大學，博士論文。