| DC 欄位 |
值 |
語言 |
| DC.contributor | 土木工程學系 | zh_TW |
| DC.creator | 廖梓淋 | zh_TW |
| DC.creator | Tzu-lin Liao | en_US |
| dc.date.accessioned | 2009-7-9T07:39:07Z | |
| dc.date.available | 2009-7-9T07:39:07Z | |
| dc.date.issued | 2009 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=963202068 | |
| 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 | 本研究主要的目的，是在利用資料填補的觀念，探討車輛偵測器資料遺失、錯誤或傳輸失真時，如何藉由上、下游偵測器資料進行填補，並經由填補績效的評估，確認上、下游偵測器能夠填補的最遠距離，藉此推算偵測器可能佈設之最大間距，以使偵測器資源能夠發揮最大的效益。本研究於資料填補分析時，首先採用K-means法將資料進行分群，而後再以類神經網路加以填補，最後並以雪山隧道偵測器所得流量、速率以及占有率三項資料進行實證分析。結果發現，兩步驟填補方式因能將同質性高的車流狀態資料匯集在一起，可獲得良好的填補績效。若合併考量流量、速率與占有率三者，則可設定佈設間距為4,900 m，此時所須容忍的平均誤差為20%（80%準確度）。若將準確度提高為90%，則佈設間距應縮短為2,100 m，但此時占有率在中低流量狀態下將無法達到此一要求。
| zh_TW |
| dc.description.abstract | In this study, we attempted to enlarge the installation spacing by using missing value treatment. Instead of densely installing vehicle detectors, we supply the traffic data on road sections in the research. We assessed the recovering performance by using a two-stage treatment method based on the traffic data. The traffic data is collected by vehicle detectors in Hshehshan tunnel. At first, the traffic data were clustered into two sets by K-means for the subsequent treatment. Second, the missing data assumed for a specified detector were recovered by neural network. It is based on traffic data detected at up- and downstream. Finally, we identified the farthest detectors whose data could still satisfy the needs of supplement at an acceptable level of accuracy. Also, the distance should be the maximum installation spacing in the circumstance.
The result shows that the method of missing value treatment used in this study can achieve good performance in general due to grouping homogeneous traffic after the preliminary clustering result. The maximum installation spacing of vehicle detectors can reach 4,900 m without losing recovering accuracy over 80%. If the spacing decreases to 2,100 m, the accuracy will even rise to 90%. However, occupancy cannot meet this requirement under low traffic flow condition.
| en_US |
| DC.subject | K-means分群 | zh_TW |
| DC.subject | 填補績效 | zh_TW |
| DC.subject | 資料填補 | zh_TW |
| DC.subject | 佈設間距 | zh_TW |
| DC.subject | 類神經網路 | zh_TW |
| DC.subject | neural network | en_US |
| DC.subject | K-means | en_US |
| DC.subject | missing value treatment | en_US |
| DC.subject | installation spacing | en_US |
| DC.title | 利用資料填補概念探討車輛偵測器佈設間距 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | A Study on the Installation Spacing of Vehicle Detectors Using the Concept of Missing Value Treatment | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |