臺灣博碩士論文加值系統

高速公路在國內南北城際運輸中扮演相當重要角色，一旦發生行車擁塞或交通事故，即可能造成嚴重之車流延滯，甚至發生追撞事故。因此，建立一套有效之事件自動偵測系統，能精確且快速地提供事件警報資訊給相關管理單位，以便進行及時之處置，實為一重要課題。基因模糊邏輯控制（GFLC）可透過自我學習之方式，尋得最佳之邏輯規則及隸屬函數組合，其精確性在許多應用領域中亦獲得相當驗證。基此，本研究嘗試利用GFLC建構一高速公路之事件自動偵測模式。
一般用以發展事件偵測系統，主要是利用車輛偵測器所偵測而得之交通即時資訊，包括流量、速度、佔有率等三種。本研究主要是利用上、下游偵測器或同一偵測器前、後時段所測得之交通資訊差作為區別，共歸納出9組變數組合。由於過多狀態變數會使得潛在邏輯規則大幅成長，導致求解精度降低之現象，故一般狀態變數不宜超過三個。因此，本研究利用GFLC建構了4個事件偵測模式，分別是流量模式、速度模式、佔有率模式及整合模式。其中，流量模式僅考慮上下游偵測器及同一偵測器之流量變數；速度模式則僅考慮其速度變數；佔有率模式則僅考慮其佔有率變數；整合模式則同時考慮不同偵測器下之流量、速度及佔有率等三種變數。此外，本研究進一步建構主成份模式，其利用主成份分析法能將9個變數經線性組合轉換成3個主成份，能有效解決GFLC之狀態變數最多三個之限制。惟建構之五個模式皆為GFLC模式，無以驗證其適用性，故本研究同時考慮此9種變數以建構類神經網路（ANN）模式並與本模式作進一步之比較分析。
為了驗證本模式之適用性，本研究主要蒐集國道一號中區路段每20秒一筆，共30個事件之交通資料進行事件偵測。並利用績效評估指標：偵測率（DR）、誤報率（FAR）與平均偵知時間（MTD）來評估模式之優劣。結果顯示，主成份模式優於其它偵測模式，得到DR=100%，FAR=0.92%及MTD=17.6秒；ANN模式得到DR=96.67%，FAR=1.43%及MTD=16.0秒；整合模式則得到DR=93.33%，FAR=1.26%及MTD=18.2秒，而其它三個僅利用單一交通資訊(流量、速度、佔有率)所建構成GFLC模式之績效，則相對較差。

Freeway system plays an important role in intercity transportation. Once traffic jams or accidents happen, the traffic would be seriously deterred and even causes other accidents. Thus, it is essential to establish an effectively incident detection system, which can provide correct and prompt alarmed incident signal to enhance the efficiency of accidents responsive actions. Genetic Fuzzy Logic Controller (GFLC) can not only self-learn the optimal combination of fuzzy rules and shapes of membership functions, but also performs very well in many applications. Based on this, the study aims to develop a GFLC-based freeway incident detection system.
The traffic information real-time detected by vehicle detection devices include volume, speed and occupancy, which are commonly used to develop incident detection system. An incident is said to be detected if a significant gap of these traffic information has been identified within different time horizons or at different vehicle detectors (upstream or downstream of the incident spot). Since the potential rules will rapidly get enlarged as the number of state variables increases, generally, the number of state variables would not exceed three. Therefore, this study develops four GFLC-based incident models based on detected traffics. They are volume model, speed model, occupancy model and integrated model, where volume model only considers the volume variables of different time horizons and vehicle detectors, speed model only considers speed variables, occupancy model only considers occupancy variables and integrated model simultaneously considers these three variables of different vehicle detectors. In addition, in order to consider as many as variables with subject to the number constraints, principal component method is used to choose three principal components, which are the linear combination of nine variables, as state variables of GFLC. This model is named as principle components model. For comparison, an artificial neural network (ANN)-based incident system is also developed, which simultaneously considers all these nine variables.
To investigate the performance and applicability of proposed models, the 20-seconds traffic data of a total of 30 accidents on national No.1. Freeway in Taiwan are collected and used to conduct a case study. Three commonly used index: detection rate (DR), false alarm rate (FAR) and mean time for detection (MTD) are adopted to measure the performances of models. The results show that the principal component model outperforms among all these incident detections models with DR=100%, FAR=0.92% and MTD=17.6 seconds, followed by ANN model with DR=96.67%, FAR=1.43% and MTD=16.0 seconds and integrated model with DR=93.33%, FAR=1.26% and MTD=18.2 seconds. Other three GFLC models based on only one kind of traffic information (volume, speed and occupancy) perform relatively inferior.

目 錄 VI
圖目錄 VIII
表目錄 IX
第一章 緒論 1
1.1研究背景與動機 1
1.2研究目的 2
1.3研究內容與流程 3
第二章 文獻回顧 6
2.1道路事件偵測之相關文獻 6
2.2基因模糊邏輯控制之相關文獻 15
2.3事件偵測演算法分類 18
2.4績效評估指標 26
第三章 研究方法 28
3.1模糊邏輯控制 28
3.1.1模糊邏輯控制之基本特性 29
3.1.2假設前提與應用限制 31
3.1.3模糊邏輯控制建構方式 32
3.2基因演算法 32
3.2.1基因演算法之基本觀念 33
3.2.2基因演算法之特性 36
3.2.3假設前提與應用限制 36
3.2.4基因演算法之操作方式 37
3.3基因模糊邏輯控制 38
3.3.1邏輯規則庫之編解碼 38
3.3.2隸屬函數之編解碼 39
3.3.3遺傳演算法則之運作 41
3.3.4 GFLC反覆演化法 43
3.4類神經網路 44
3.4.1神經網路架構 44
3.4.2倒傳遞網路之基本概念 46
3.4.3倒傳遞類神經網路 47
第四章 模式建構 51
4.1變數選擇 51
4.2事件偵測模式 52
4.3 GFLC事件偵測模式 56
4.3.1邏輯規則與隸屬函數之編解碼 56
4.3.2遺傳演算法則之運作 56
4.3.3 GFLC反覆演化機制 58
4.3.4適合度值 58
第五章 實例應用 60
5.1資料整理與特性分析 60
5.1.1事件資料整理 60
5.1.2事件資料特性分析 62
5.2績效評估 66
5.3模式參數分析及設定 69
5.4學習結果分析 70
第六章 GFLC與ANN之比較 75
6.1 ANN之建構 75
6.2 ANN模式之訓練及驗證 76
6.3 GFLC與ANN之比較分析 79
第七章 結論與建議 81
7.1結論 81
7.2建議 82
參考文獻 84

1.中文部分
【1】王秀帆，民92年，「應用灰色預測於高速公路事件自動偵測之研究」，國立中央大學土木工程學所碩士論文。
【2】王雲慶，民94年，「應用基因模糊邏輯建構適應性匝道儀控系統」，逢甲大學交通工程與管理學系碩士班碩士論文。
【3】方守傑，民93年，「鑑定專家系統之建構與驗證」，逢甲大學交通工程與管理學系碩士班碩士論文。
【4】周義華、許鉅秉、李啟仲，民89年，「高速公路事故屬性即時自動鑑別之方法研究」，運輸計劃季刊，第二十九卷第三期， pp.499~528。
【5】邱裕鈞、藍武王，民90年，「應用遺傳演算法建構適應性模糊邏輯控制系統─以跟車行為為例」，中華民國運輸學會第16屆學術論文研討會，第515-526頁。
【6】侯鈞元，民91年，「應用羅吉特模式於市區道路事故偵測系統」，國立成功大學交通管理科學研究所碩士論文。
【7】楊雨青，民88年，「高速公路事件偵測與匝道儀控整合模式之研究-類神經網路之應用」，國立成功大學碩士論文。
【8】黃振賢，民81年，「高速公路事件自動偵測方法之研究」，國立中央大學土木工程學所碩士論文。
【9】曾信忠，民82年，「模糊理論應用於高速公路事件自動偵測之研究」，國立中央大學土木工程學所碩士論文。
【10】馮正民、邱裕鈞，民93年，「研究分析方法」，建都文化事業股份有限公司。
【11】彭國倫，民90年，「Fortran 95程式設計」，碁峰資訊股份有限公司。
【12】葉源祥，民84年，「高速公路事件偵測之微觀車流參數法」，國立台灣大學土木工程學研究所碩士論文。
【13】賴建華，民92年，「適應性基因模糊邏輯號誌控制系統」，國立交通大學交通運輸研究所碩士論文。
【14】魏健宏、黃國平、陳昭宏，民85年，「應用人工神經網路發展高速公路意外事件自動偵測模式」，運輸計劃季刊，第二十五卷第二期，pp.209~234。
2.英文部分
【1】Ai-Deek, H.M., Ishak, S.S. and Khan, A.A., “Impact of freeway geometric and incident characteristics on incident detection,” Journal of Transportation Engineering, Vol.122, No.6, pp.440-446, 1996.
【2】Adeli, H. and Karim, A., “Fuzzy-wavelet RBFNN model for freeway incident detection,” Journal of Transportation Engineering, Vol.126, No.6, pp.464-471, 2000.
【3】Abdulhai, B. and Ritchie, S.G., “Enhancing the universality and transferability of freeway incident detection using a Bayesian-based neural network” Transportation Research Part C7, pp261-280, 1999.
【4】Cheu, R.L. and Ritchie, S.G., “Automatic Detection of Lane-Blocking Freeway Incident Using Artificial Neural Networks”, Transportation Research Part C, Vol.3 , NO.6, pp.371-388, 1995.
【5】Chan, P.T., Xie, W.F. and Rad, A.B., “Tuning of fuzzy controller for an open-loop unstable system: a genetic approach” Fuzzy Sets and Systems, 111, pp.137-152, 2000.
【6】Chiou, Y.C. and Lan, L.W., “Genetic fuzzy logic controller: An iterative evolutionalgorithm with new encoding method,” Fuzzy Sets and Systems, Vol.152, Issue 3, pp.617-635, 2005.
【7】Dia, H. and Rose, G.., “Development and evaluation of neural network freeway incident detection models using field data,” Transportation Research Part C , Vol.5 , No.5 , pp.313-331, 1997.
【8】Homaifar, A. and McCormick, E., “Simultaneous design of membership functions and rule sets for fuzzy controllers using genetic algorithms” IEEE Transactions on fuzzy systems, Vol.3, No.2, pp.129-139, 1995.
【9】Ishak, S. and AI-Deek H., “Performance of automatic ANN-Based incident detection on freeways” Journal of Transportation Engineering, Vol.125, No. 4, pp.281-290, 1999.
【10】Khan, S.I. and Ritchie, S.G., “Statistical and neural classifiers to detect traffic operational problems on urban arterials” Transportation Research Part C6 pp.291-314, 1998.
【11】Karim, A. and Adeli, H., “Comparison of fuzzy-wavelet radial basis function neural network freeway incident detection model with California algorithm” Journal of Transportation Engineering, Vol.128, No. 1, pp.21-30, 2002.
【12】Khoury, J.A., Haas, C.T., Mahmassani, H., Logman, H. and Rioux, T., “Performance caparison of Automatic vehicle identification and inductive loop traffic detectors for incident detection” Journal of Transportation Engineering, Vol.129, No. 6, pp.600-607, 2003.
【13】Lan, L.W. and Huang, Y.C., “A rolling-trained fuzzy neural network（RTFNN）approach for freeway incident detections”, 2005.
【14】Mak, C.L. and Henry S.L., “Transferability of expressway incident detection algorithms to Singapore and Melbourne” Journal of Transportation Engineering, Vol.131, No. 2, pp.101-111, 2005.
【15】Shieh, C.S., “Genetic fuzzy control for time-varying delayed uncertain systems with a robust stability,” Applied Mathematics and Computation, 131, pp.39-58, 2002.
【16】Sheu, J.B., “A fuzzy clustering –based approach to automatic freeway incident detection and characterization” Fuzzy Sets and Systems, 128, pp.377-388, 2002.
【17】Thrift, P., “Fuzzy logic synthesis with genetic algorithms,” Proceeding of the Fourth International Conference on Genetic Algorithms, pp.509-513, 1991.
【18】Xiong, N. and Litz, L., “Reduction of fuzzy control rules by mans of premise learning-method and case study,” Fuzzy sets and systems, 132, pp.217-231, 2002.