跳到主要內容

臺灣博碩士論文加值系統

(3.229.117.123) 您好!臺灣時間:2022/08/12 16:12
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:林逸塵
研究生(外文):Yat-Chen Lin
論文名稱:類神經網路應用於空氣品質預測之研究
論文名稱(外文):Application of Artificial Neural Network on The Prediction of Ambient Air Quality
指導教授:袁中新袁中新引用關係
學位類別:碩士
校院名稱:國立中山大學
系所名稱:環境工程研究所
學門:工程學門
學類:環境工程學類
論文種類:學術論文
論文出版年:2002
畢業學年度:90
語文別:中文
論文頁數:133
中文關鍵詞:能見度類神經網路空氣品質懸浮微粒
外文關鍵詞:visibilityartificial neural networkair qualityPM10
相關次數:
  • 被引用被引用:94
  • 點閱點閱:1958
  • 評分評分:
  • 下載下載:611
  • 收藏至我的研究室書目清單書目收藏:0
高屏空品區為全台灣空氣品質最差之地區,高雄市空氣污染事件日皆為懸浮微粒及臭氧所造成,其中因懸浮微粒濃度過高所造成之事件日站日數更佔總事件日數一半以上。此外,高雄市空氣品質不良,除可藉由空氣品質指標加以判斷外,亦反映於大氣能見度不良之情形。由於能見度可被民眾感官所直接感受,因此能見度不佳時常為民眾所詬病。大氣能見度不良除受天氣型態之影響外,與空氣污染物濃度有一定程度之相關性,其中懸浮微粒之消光效應為造成能見度降低之主要因素。
本研究擬針對高雄都會區之懸浮微粒濃度及能見度,以類神經網路加以預測,本研究之目的有二:(1)針對氣象因子及空氣污染物濃度對懸浮微粒濃度及能見度之影響加以探討與分析。(2)利用類神經網路預測懸浮微粒濃度與大氣能見度,並討論類神經網路之優缺點及可能之改善方法。
懸浮微粒部份,本研究嘗試將夏季及冬季數據分開進行網路測試,藉以瞭解分季與不分季對於網路預測結果之可能影響。研究結果顯示,懸浮微粒全年不分季網路測試最佳輸入變數組合為PM10、氣壓、全天空輻射量、相對溼度、氣溫及雲量等六項,所得之網路測試結果最佳,網路測試值與實際懸浮微粒濃度間之相關係數呈高度相關(R=0.876)。而就夏、冬分季網路預測結果而言,則以夏季之網路測試結果較佳,網路測試值與實際值間呈高度相關(R=0.753),惟較不分季網路結果略低;而冬季網路測試結果最差,與實際值僅呈中度相關(R=0.553)。
能見度網路測試分為兩階段進行,第一階段(set-1 ~ set-3)網路之研究結果顯示,氣象變數中以相對溼度、氣溫、雲量影響較大,污染物方面則以PM10、O3、NO2較具影響,網路測試結果以輸入變數僅含氣象變數之組set-1最差(R=0.586),而以輸入變數含污染物濃度之set-3結果較佳,惟僅與實際值呈中度相關(R=0.633)。第二階段(set-4、set-5)針對隱藏層神經元數及輸入變數加以調整,並加入能見度作為網路輸入變數,網路測試之結果與實際值間之相關性並無提昇,但由圖形趨勢及極值、平均值等統計資料觀之,則有明顯改善之現象。
本研究最後以網路進行實際之短期預測,並以PSI及能見度分級進行試預報,結果顯示PM10不分季及夏季兩組網路之預報分級準確性為76.9%。能見度方面,以三級制及五級制預報之結果而言,第一階段之set-3正確率同為76.9%最佳。
The air quality in Kaohsiung and Ping-Dong district is the worst in Taiwan. The air pollution episodes in Kaohsiung are attributed to high concentrations of PM10 and O3. Among them, over half of the episodes result from PM10. In addition to Pollutant Standards Index (PSI), atmospheric visibility is also an indicator of ambient air quality. Citizens always complain about the impairment of visibility because it can be visualized directly. Visibility is closely correlated to both air pollutants and meteorological condition. Extinction of visible light by fine particles is the major reason for visibility impairment.
In this study, an artificial neural network was applied to predict the concentration of PM10 and atmospheric visibility. The objectives of this study were to investigate the effects of meteorological factor and air pollutants on visibility and to apply artificial neural network to predict the concentration of PM10 and atmospheric visibility.
The measured PM10 data were divided into two parts (i.e. summer and winter, ) to understand whether different season affect the prediction of PM10 concentration. The modeling results showed that the optimum input variables included the PM10 concentration, atmospheric pressure, surface radiation, relative humidity, atmospheric temperature, and cloud condition. The network outputs showed high correlation with measured PM10 concentration (R=0.876) in the whole-year set. Furthermore, the prediction of summer set also showed high correlation with measured PM10 concentration (R=0.753). The winter set demonstrated the worse prediction among three sets, and showed medium correlation with measured PM10 concentration (R=0.553).
The visibility network test was conducted by two stages. The first stage (set-1~set-3) showed that relative humidity, atmospheric temperature, and cloud condition were the most important meteorological factors, while PM10, O3, and NO3 were the most important air pollutants on the prediction of atmospheric visibility. The prediction of set-1 considering only meteorological factors was the worst (R=0.586), while set-3 was the best and showed medium correlation with measured atmospheric visibility (R=0.633). The second stage (set-4 and set-5) increased the hidden neuron numbers and input variables, and added atmospheric visibility in the input variables. Although the correlation coefficients between predicted and measured data did not increase, the prediction of atmospheric visibility had significant improvement.
Finally, a short-term prediction of PM10 and atmospheric visibility was conducted and validated by the level of PSI values and atmospheric visibility. Prediction results showed that the accuracy of PM10 prediction was 76.9%, while the prediction of atmospheric visibility by set-3 network demonstrated an accuracy of 76.9%. Moreover, no significant difference of prediction was detected by using either three-level or five-level visibility systems.
目 錄
謝誌…………………………………………………………………..I
中文摘要…………………………………………………………….II
英文摘要…………………………………………………………….IV
目錄………………………………………………………………….VI
表目錄……………………………………………………………….IX
圖目錄……………………………………………………………….XII
第一章 前言……………………………………………………….1-1
1-1 研究緣起…………………………………………………1-1
1-2 研究目的…………………………………………………1-2
第二章 文獻回顧………………………………………………….2-1
2-1 高雄地區氣象及空氣品質現況描述……………………2-1
2-2 影響PM10及能見度之相關變數探討……………………2-13
2-2-1 影響PM10之變數探討……………………………2-13
2-2-2 影響能見度之變數探討………………………….2-18
2-3 不同類型預測模式探討比較……………………………2-20
2-3-1 傳統計量迴歸模式……………………………….2-20
2-3-2 專家系統………………………………………….2-21
2-3-3 模糊邏輯理論…………………………………….2-21
2-3-4 遺傳演算法……………………………………….2-22
2-3-5 類神經網路……………………………………….2-23
第三章 類神經網路原理與應用………………………………….3-1
3-1 類神經網路原理…………………………………………3-1
3-1-1 類神經網路簡介………………………………….3-1
3-1-2 類神經網路模式架構…………………………….3-3
3-1-3 類神經網路種類………………………………….3-6
3-2 倒傳遞類神經網路………………………………………3-11
3-3 類神經網路在空氣品質預測之應用……………………3-14
第四章 研究方法………………………………………………….4-1
4-1 數據前處理………………………………………………4-1
4-2 網路參數設定原則………………………………………4-2
4-3 最佳輸入變數組合篩選…………………………………4-4
4-4 敏感度分析………………………………………………4-6
4-5 網路輸出結果判斷原則…………………………………4-6
4-6 研究流程……………………………..……………………4-7
第五章 結果與討論……………………………………………….5-1
5-1 網路參數數設定……………….…………………………5-1
5-2 懸浮微粒濃度網路訓練與測試………..…………………5-2
5-2-1 懸浮微粒濃度網路訓練….…………………….….5-5
5-2-2 懸浮微粒濃度網路測試結果比較………………5-16
5-3 能見度網路訓練與測試……..……………………….…..5-23
5-3-1 能見度網路訓練….……………………………….5-24
5-3-2 網路效能改善測試….…………………………….5-32
5-3-3 能見度網路測試結果比較……………………….5-36
5-3-4 降雨量數據測試………………………………….5-43
5-4 空氣品質預報……………………………………………5-45
5-4-1 懸浮微粒濃度預報………………………………..5-45
5-4-2 能見度預報………………………………………...5-52
第六章 結論與建議……………………………………………….6-1
6-1 結論……………………………………………….….……6-1
6-2 建議………………………………………………….….…6-3
參考文獻………………………………………………………………7-1
中興工程顧問公司,「高雄市空氣品質管理計劃」,高雄市政府環境保護局,2000。江世民,「台北地區臭氧及懸浮微粒型態與其他污染物及氣象條件之分析比較」,第十六屆空氣污染控制技術研討會,1999。吳清吉、王嘉宏,「氣象條件對空氣品質的影響」,第七屆空氣污染控制技術研討會,1990。吳敏生,「模糊理論在時間延遲系統上之研究」,正修學報第11期,1998。李謨中,「類神經網路應用於飛機結構負載監控模擬系統之研究」,國立成功大學,博士論文,1998。林斌龍,李崇德,「台北地區PM10濃度與氣象因子之多變量統計分析」,第七屆空氣污染控制技術研討會,1990。林碧亮,廖述良,朱佳仁,「以時間序列法預測空氣品質之初步研究」,第十一屆空氣污染控制技術研討會,1994。林俊賢,「廣義類神經網路與資訊理論在化工程序的應用」,國立清華大學,博士論文,1998。洪崇恩,「以類神經網路預測台灣股價報酬率-以電子股為例」,朝陽大學,碩士論文,1999。陳清暉,「類神經網路結合遺傳演算法應用於工業區廢水處理廠放流水品質預測之研究」,雲林科技大學,碩士論文,1998。袁中新,楊宏宇,林文印,袁菁,周本生,李崇垓,「空氣中能見度改善計劃」,高雄市政府環境保護局,1999。袁中新,楊宏宇,袁菁,羅金翔,「能見度為指標之空氣品質改善暨提升計劃」,高雄市政府環境保護局,2000。翁誌煌,袁菁,袁中新,李崇垓,林逸塵,「高雄市空氣污染物指標研析」,高雄市政府環境保護局,2001。莊秉潔,「地表層擴散係數、風速、雨量及混合層高對台北都會早上八時懸浮微粒濃度之影響」,中國環境工程學刊,第三卷,第二期,1993。曹勝雄,曾國雄,江勁毅,「傳統計量迴歸、模糊迴歸、GMDH、類神經網路四種方法在預測應用上之比較-以國人赴港旅客需求之預測為例」,中國統計學報第34卷第2期,1996,pp.132-161。莊源鍵,「類神經網路應用於一般廢棄物焚化廠煙道氣品質預測之研究」,國立雲林科技大學,碩士論文,1998。張順欽,李崇德,陳稼興,「以類神經網路預測台北地區臭氧濃度之預測」,第二屆環境系統分析研討會,1999。張順欽,李崇德,「大陸沙塵影響空氣中懸浮微粒濃度之評估」,第九屆中華民國氣膠科技研討會,2001。許純君 譯,Stephen A. DeLurgio 原著,「預測的原則與應用第一版」,1999。莊秉潔,許家彰,陳建隆,「懸浮微粒濃度與微氣象因子之關係探討-台北案例研究」,第十八屆空氣污染控制技術研討會,2001。黃尚雄,「應用類神經網路預測空氣品質之研究」,國立交通大學,碩士論文,1995。程萬里,黃怡嘉,「中部地區臭氧及懸浮微粒與氣象因子之相關研究」,第十七屆空氣污染控制技術研討會,2000。楊敏生,「模糊理論簡介」,數學傳播第18卷第1期,1994。萬騰州,張乙森,崔天佑,「應用類神經網路模式於台灣地區廢水處理廠成本函數之研究」,第二屆環境系統分析研討會,1999。萬騰州,崔天佑,陳清暉,「類神經網路結合遺傳演算法應用於工業區聯合廢水處理場放流水品質預測之研究」,第二屆環境系統分析研討會,1999。萬騰州,范惠婷,許家宴,方景萱,「應用類神經網路於雲林地區空氣品質(PM10)預測之研究」,雲林科技大學環境與安全工程系,實務專題,2001。葉怡成,「類神經網路模式應用與實作第七版」,2000。鄭淑燕,望熙榮,鄭曼婷,「台中地區空氣品質分析」,第八屆空氣污染控制技術研討會,1991。劉山豪,「高雄都會區消光係數與能見度量測及細微粒污染源貢獻量解析」,國立中山大學,碩士論文,2000。盧炳志,「類神經網路在大地工程參數分析之研究」,國立成功大學,博士論文,1998。盧瑞山,「類神經網路於環境資訊之鑑視、推估及預測之研究」,國立台灣大學,博士論文,1998。盧瑞山,駱尚廉,「類神經網路於環境工程上之應用分析與回顧」,第十二屆環境規劃與管理研討會,1999。羅華強,「類神經網路-MATLAB的應用第六版」,2001。蘇侯洵,林昱成,白曛綾,「以三種時間序列法預估大氣中臭氧濃度較」,第十一空氣污染控制技術研討會,1994。Acuna G., Jorquera H., Perez R., ”Neural network model for maximum ozone concentration prediction., ”Lecture Notes in Computer Science, No.1112, pp.263-268, 1996.Baik, Nam-Jun, Yong Pyo Kim, Kil Choo Moon, ”Visibility Study in Seoul, 1993, ”Atmospheric Environment, Vol.30, pp.2319-2328, 1996.Bloomfiled, Peter, J. Andrew Royle, Laura J. Steinberg, Qing Yang, ”Accounting For Meteorological Effects In Measuring Urban Ozone Levels And Trends, ”Atmospheric Environment, Vol.30, No.17, pp.3067-3077, 1996.Boznar, Marija, Martin Lesjak and Primoz Mlakar, ”A Neural Network-Based Method for Short-Term Prediction of Ambient SO2 Concentrations in highly Polluted Industrial Areas of Complex Terrain, ”Atmospheric Environment, Vol. 27B, No.2, pp.221-230, 1993.Comrie, A. C., ”Comparing neural networks and regression models for ozone forecasting, ”Journal of Air and Waste Management, Vol.47, pp.653-663, 1997.Deacon, Andrew R., Richard G. Derwent, Roy M. Harrison, Doug R. Middleton, Steve Moorcroft, ”Analysis and interpretation of measurements of suspended particulate matter at urban background sites in the United Kingdom, ”The Science of the Total Environment, Vol. 203, pp.17-36, 1997.Gardner, M. W. and Dorling, S. R., ”Artificial neural networks (The multiplayer perceptron )-A review of applications in the atmospheric sciences, ”Atmospheric Environment, Vol.32, pp.2627-2636, 1998.Gardner, M. W. and Dorling, S. R., ”Neural network modeling and prediction of hourly NOX and NO2 concentrations in urban air in London, ”Atmospheric Environment, Vol.33, pp.709-719, 1999.Lai, Lee Yuet, R. Sequeira, ”Visibility degradation across Hong Kong:its components and their relative contributions, ”Atmospheric Environment, Vol.35, pp.5861-5872, 2001.Lai, Lee Yuet, R. Sequeira, ”Water-soluble aerosol and visibility degradation in Hong Kong during autumn and early winter, 1998, ”Environmental Pollution, Vol.116, pp.225-233, 2002.Lam K. C., Cheng S., ”A synoptic climatological approach to forecast concentrations of sulfur dioxide and nitrogen oxides in Hong Kong, ”Environmental Pollution, Vol.101, pp.183-191, 1998.Lam K. C., Cheng S., ”Synoptic typing and its application to the assessment of climatic impact on concentrations of sulfur dioxide and nitrogen oxides in Hong Kong, ”Atmospheric Environment, Vol.34, pp.585-594, 2000.MacDonald, Clinton P., Paul T. Roberts, Hilary H. Main, Timothy S. Dye, Dana L. Coe, James Yarbrough, ”The 1996 Paso del Norte Ozone Study:analysis of meteorological and air quality data that influence local ozone concentrations, ”The Science of the Total Environment, Vol.276, pp.93-109, 2001.Macpherson, K. P., A. J. Conway, J. C. Brown, ”Prediction of solar and geomagnetic-activity data using neural networks, ”Journal of Geophysical Research-Space Physics, Vol.100, No.A11, pp.21735-21744.Marko, K. A. et al., ”Automotive diagnostics using trainable classifiers:statistical testing and paradigm selection, ”IJCNN-90-San. I, pp.33-38, 1990.Marose, R. A., ”A financial neural-network application, ”AI EXPERT, pp.50-59, 1990.Monn, C.H., O.Braendli, G.Schaeppi, C.H.Schindler, U.Ackermann, P.H.Leuenberger and T.Sapldia, ”Particulate Matter < 10 mm in urban , Rural and Alpine Air in Switzerland, ”Atmospheric Envernment, Vol.29, pp.2565-2573, 1995.Moseholm, L., J. Silva, T. Larson, ”Forecasting carbon monoxide concentrations near a sheltered interection using video traffic surveillance and neural networks, ”Transportation Research 1D, pp.15-28, 1996.Muxworthy, Adrian R., Jurgen Matzka, Nikolai Petersen, ”Comparison of magnetic parameters of urban atmospheric particulate matter with pollution and meteorological data, ”Atmospheric Environment, Vol.35, pp.4379-4386, 2001.Peton, N., G. Dray, D. Pearson, M. Mesbah, B. Vuillot, ”Modelling and analysis of ozone episodes, ”Environmental Modelling & Software, Vol. 15, pp.647—652, 2000.Podnar, Domagoj, Darko Koracin, Anna Panorska, ”Application of artificial networks to modeling the transport and dispersion of tracers in complex terrain, ”Atmospheric Environment, Vol.36, pp.561-570, 2002.Poissant, Laurier, Jan W. Bottenheim, Pascal Roussel, Neville W. Reid, Hiromi Niki, ”Multivariate Analysis of a 1992 SONTOS Data Subset, ”Atmospheric Environment, Vol.30, No.12, pp.2133-2144, 1996.Rege, A. R., R. W. Tock, ”A simple neural network for estimating emission rates of hydrogen sulfide and ammonia from single point sources, ”Journal of Air and Waste Management Association, Vol.46, pp.953-962, 1996.Reich, S. L., D. R. Gomez, L. E. Dawidowski, ”Artificial neural network for the indentification of unknown air pollution sources, ” Atmospheric Environment, Vol.33, pp.3045-3052, 1999.Robeson S. M., Steyn D. G., ”Evaluation and comparison of statistical forecast models for daily maximum ozone concentration, ” Atmospheric Environment, Vol.24B, pp.303-312, 1990.Rutllant, Jose, Rene Garreaud, ”Meteorological Air Pollution Potential For Santiago, Chile:Towards An Objective Episode Forecasting, ” Environmental Monitoring and Assessment, Vol.34, pp.223-244, 1995.Sequeira, R., Kim-Hung Lai, ”Short communication the effect of meteorological parameters and aerosol constituents on visibility in urban Hong Kong, ”Atmospheric Environment, Vol.32, pp.2865-2871, 1998.Sohn, Sang Hyum, Oh, Sea Cheon, Jo, Byung Wan, and Yeo, Yeong-Koo, “Prediction of Ozone Formation Based on Neural Network, ” Journal of Environmental Engineering, pp.688-696, 2000.Smith, Steve, F.Trevor Stribley, Peter Milligan, Ben Barratt, ”Factors influencing measurements of PM10 during 1995-1997 in London, ” Atmospheric Envernment, Vol.35, pp.4651-4662, 2001.Song, Xin-Hua and Philip K. Hopke, ”Solving the chemical mass balance problem using an artificial neural network, ”Environ. Sci. Technol, Vol. 30, pp.531-535, 1996.Tsuang B.-J., Chao C.-P., ”Application of circute model for Taipei City PM10 simulation, ”Atmospheric Envernment, Vol.33, pp.1789-1801, 1999.Van Der Wal, J.T., L.H.J.M. Janssen, ”How contribute emission of PM10 and meteorology to concentrations of fine particles in the Netherlands, ”J.Aerosol Sci., Vol.27, Suppl.1, pp.S681-S682, 1996.Vega, E., E. Reyes, G. Sanchez, E. Ortiz, M. Ruiz, J. Chow, J. Watson, S. Edgerton, ”Basic statistics of PM2.5 and PM10 in the atmosphere of Mexico city, ”The Science of Total Environment, Vol.287, pp.167-176, 2002.Verdecchia, M., G. Visconti, F. D’Andrea, S. Tibaldi, ”A neural network approach for blocking recognition, ”Geophysical Research Letters, Vol.23, No.16, pp.2081-2084, 1996.Yi, J. and Prybutok, R., ”A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area, ”Environmental Pollution, Vol.92, pp.349-357, 1996.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊