| DC 欄位 |
值 |
語言 |
| DC.contributor | 土木工程學系 | zh_TW |
| DC.creator | 蕭銓宏 | zh_TW |
| DC.creator | Chuan-Hung Hsiao | en_US |
| dc.date.accessioned | 2006-7-19T07:39:07Z | |
| dc.date.available | 2006-7-19T07:39:07Z | |
| dc.date.issued | 2006 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:444/thesis/view_etd.asp?URN=93322067 | |
| dc.contributor.department | 土木工程學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 由於長隧道密閉式的空間特性，一旦事故於隧道內發生，極易釀成二次事故。目前國內隧道事故偵測方式，是先由用路人回報，再由附近的閉路電視確認是否真的有事故發生，此法雖然能夠準確的得知事故發生的地點及事故型態，但耗時甚鉅，在這時間內可能已造成相當大的人員及財務損失，因此，發展有效的長隧道事件偵測系統是不可或缺的。
本研究的目的在於發展一套迅速且準確的隧道自動事件偵測演算法，用來提供交通管理者事件發生時的相關資訊，讓事件所造成的損失減到最小。本研究利用自動車輛辨識技術獲得微觀交通參數資料，構建微觀車流參數法，再以車流模擬的方式模擬於長隧道出入口各架設一座影像式自動車輛辨識系統，以此方式獲得構建演算法微觀車流參數，最後以三個績效評估因子：偵測率、平均偵測時間及誤報率來評估隧道自動事件偵測演算法。
經由一系列的評估分析，整理得本研究所提出的事件偵測演算法其偵測率於高中低流量狀態下均超過90%，誤報率均小於4%，平均偵測時間約為100秒。本研究所提出的事件偵測演算法具有高偵測率、偵測時間短的優異表現，在誤報率方面也屬可接受的表現，因此本研究的結果確可提供相關單位研究或建置系統參考用。 | zh_TW |
| dc.description.abstract | Because of the characteristic of the long airtight type of tunnel, once an incident occurs in the tunnel, a secondary incident can be induced further. The procedure of tunnel incident detection are: (1)emergency call from the passerby, (2)and then confirm it by nearby closed-circuit television. Using this method we can detect the location and type of incident, but this procedure isn’t effective because it take too much time to detect and may result in great loss and injury. Therefore, it is indispensable to develop a long tunnel incident detection system.
This research focus on developing a rapid and correct automatic tunnel incident detection algorithm to minimize the damages from incidents. We use microscopic traffic parameters based on automatic vehicle identification technology to construct this tunnel incident detection algorithm. Because of automatic vehicle identification facility not being constructed in the tunnel yet, this research verifies the performance of the incident detection algorithm by utilizing traffic simulation. Then we evaluate the performance with three factor, including detection rate, mean detective time, and false alarm rate.
After series analysis of the incident algorithm, it is concluded that the detection rate is more than 90%, false alarm rate is less than 4%, and mean detective time is approximately 100 second. It is shown that performance of this research is superior in detection rate and mean detective time. And the performance in false alarm rate is also acceptable. Therefore, the algorithm developed by this research can be provided to other researches and relative associations for reference. | en_US |
| DC.subject | 自動事件偵測 | zh_TW |
| DC.subject | 自動車輛辨識 | zh_TW |
| DC.subject | 長隧道 | zh_TW |
| DC.subject | automatic vehicle identification | en_US |
| DC.subject | long tunnel | en_US |
| DC.subject | automatic incident detection | en_US |
| DC.title | 應用自動車輛辨識技術於長隧道自動事件偵測 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | A Algorithm for Urban Tunnel Automatic Incident Detection with Automatic Vehicle Identification Technology | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |