Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/10165
標題: 以適應性網路架構模糊推論系統預測工程建設垂直變動量之研究-以路基與建物為例
A Study of Forecasting Engineering Construction Vertical Deformation by Adaptive Network-based Fuzzy Inference System – Case Study of Roadbed and Building
作者: 韓孟辰
Han, Mon-Chen
關鍵字: 適應性網路架構模糊推論系統(ANFIS);Adaptive Network-based Fuzzy Inference System (ANFIS);沉陷預測;project roadbed settlement
出版社: 土木工程學系所
引用: 1. 王士金,2005,時間序列及貝葉斯模型在沉降預測中的應用,浙江大學市政工程學系碩士論文。 2. 石磊,2012,淺談路基沉降預測與計算方法,工程技術,第六期,第115頁。 3. 朱紅霞,2004,神經網絡模型在高速公路軟基沉降預測中的應用,天津大學岩土工程學系碩士論文。 4. 李磊,2004,地基沉降預測方法分析,浙江大學岩土工程學系碩士論文。 5. 吳東軒,2005,精密水準測量誤差分析:轉點沉陷效應之探討,國立成功大學測量及空間資訊學系碩士論文 6. 徐華騏,2008,柔性計算(類神經網路與模糊集理論)用於颱洪流量預測-以五堵集水區為例,清雲科技大學空間資訊與防災科技研究所碩士論文。 7. 崔海麗,2009,高速公路軟土地基沉降預測方法的研究及其應用,山東大學結構工程學系碩士論文。 8. 陳建谷,2003,應用倒傳遞類神經及適應性模糊類神經網路模式預測垃圾焚化廠煙道器之比較研究,國立雲林科技大學環境與安全工程所碩士論文。 9. 陳善雄、王星運、許錫昌、余飛、秦尚林,2011,路基沉降預測的三點修正指數曲線法,岩土力學,第十一期,第3355頁至3360頁。 10. 陳金良,2002,感應電動器之模糊類神經網路控制器之研究,國立成功大學工程科學系碩士論文。 11. 陶雨龍,2004,改良式模糊類神經網路與其應用,私立中原大學工業工程學系碩士論文。 12. 裴善成,2007,應用強化式學習建構模糊類神經控制系統,國立中山大學電機工程系碩士論文。 13. 管晏如,2000,測量學,友寧出版社。 14. 蔡銘銓,2010,Logistic與Gompertz曲線模型擬合在建物沉降分析之研究,國立中興大學土木工程學系碩士論文。 15. 鄭遠鐘,2000,適應性類神經模糊控制器於泵浦系統之應用,國立中央大學機械工程學所碩士論文。 16. 劉全,2004,高速公路軟土路基沉降預測方法研究,河海大學港口海岸與近海工程學系碩士論文。 17. 劉伏成,2004,路堤軟基沉降預測方法研究,武漢理工大學橋梁與隧道工程學系碩士論文。 18. 羅竣文,2010,應用類神經網路於颱風降雨量及時預報之研究,國立台灣大學土木工程學系碩士論文。 19. L.A.Zadeh,1965., “Fuzzy set”, Information and Control, Vol.8,pp.338-358. 20. Jang, J.S.R.,1993.,ANFIS:Adaptive-network-based fuzzy inference system. IEEE Transactions on Systems, Man, and Cybernetics, v 23, n 3,p 665-685. 21. Mamdani,E.H.;Assilian,S,1975.,Experiment In Linguistic Synthesis With A Fuzzy Logic Controller. International Journal of Man-Machine Studies, v 7, n 1, p 1-13. 22. Takagi, Tomohiro ; Sugeno, Michio1985.,Fuzzy Identification Of Systems And Its Applications To Modeling And Control, IEEE Transactions. on Systems, Man, and Cybernetics, v SMC-15, n 1, p.116-132.
摘要: 
為了促進經濟的增長,政府於60年代末期推動十大建設,為台灣往後的經濟奇蹟打下穩固的基礎,也使得國家進入了大興土木的時期,然而台灣位處於板塊交界處且氣候多變,地質並不是很優良,若土地過度開發,將使得脆弱的地質更顯得雪上加霜。工程建設若建立於脆弱的地質上,勢必將產生沉陷。為了規劃工程施工進度與結構物之穩定度,則必須對工程設施進行沉陷量之觀測。
沉陷的類型可分為施工沉陷與施工後沉陷,前者總沉陷量較大且具有規律性,後者總沉陷量小且具有隨機性。本研究以路基沉陷做為施工沉陷範例,以電廠沉陷做為施工後沉陷範例,利用適應性網路架構模糊推論系統建立沉陷模型,並進行相關之分析與研究。
研究成果顯示,以適應性網路架構模糊推論系統建立之沉陷模型,能夠準確地預測路基沉陷量,最佳之預測精度可達到±0.4公厘,若精度要求較低時,最多可以減少37.6%之訓練樣本。而在預測電廠沉陷量方面,精度可達到±1.16公厘與±1.19公厘,由於電廠總沉陷量小,因此相對誤差高達22.18%與24.69%。本研究之成果有助於工程之設計與決策,並可做為工程建設進行補強之依據,以增加工程結構物之穩定度及使用年限。

For advancing the economy to develop. The government promoted Ten Major Construction Projects in the 1960s, laying the firm foundation for Taiwan’s economic miracle. It also makes the country entered a period of building many constructions. However, Taiwan is located at the junction of plates and the climate is very changeable. Geology is not very good. If the land resources are over-development, that will make the fragile geology being worse. Structure built on the fragile geology that is bound to produce vertical deformations. For planning the works progress and structural stability. It has to monitor the vertical deformations of the structures.
There are two types of vertical deformations:Building vertical deformations and after building vertical deformations. The former has larger total vertical deformations with regularity. The latter has littler total vertical deformations with randomness. In this study, it takes roadbeds’ vertical deformations for building vertical deformations’ example, and power plants’ vertical deformations for after building vertical deformations. Use Adaptive Network-based Fuzzy Inference System(ANFIS) to establish vertical deformations’ models, and progress related analysis and research.
According to the researchable results, using ANFIS to establish vertical deformations’ models that can precise project roadbeds’ vertical deformations. The best precision can achieve ±0.4mm. If requirement of precision is lower, the training data can be reduce 37.6%. In terms of power plants’ vertical deformations, the precision can achieve ±1.16mm and ±1.19mm. Because power plants’ total vertical deformations are little, the relative errors are up to 22.18% and 24.69%. The results of this study contribute to design and decision in engineering. And it can be a support to reinforce the structures to add using years of structures.
URI: http://hdl.handle.net/11455/10165
其他識別: U0005-1008201202251800
Appears in Collections:土木工程學系所

Files in This Item:
File SizeFormat Existing users please Login
nchu-101-7099062071-1.pdf1.49 MBAdobe PDFThis file is only available in the university internal network    Request a copy
Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.