Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/49629
Title: | 都市地區控制測量成果建置與維護策略 On the Establishment and Maintenance of Control Surveys in Urban Areas |
Authors: | Cheng-Wei Yen 顏誠緯 |
Advisor: | 韓仁毓(Jen-Yu Han) |
Keyword: | 控制測量,網形平差,虛擬序列式平差,不變觀測量,測繪成果維護, Control Surveys,Adjustment of Networks,Unified Sequential Adjustment,Invariant Observations,Maintenance of Control Surveys, |
Publication Year : | 2016 |
Degree: | 碩士 |
Abstract: | 政府機關為便於管理土地,爰施行地籍測量以繪製地籍圖界定土地權利範圍,而為了產製高精度圖資以保障人民權益,首要係建立精良之控制測量成果。然而都市地區控制測量成果常因頻繁公共建設導致地面點位滅失,且歷經不同年代的補建、不同的坐標系統轉換及不同測設方法等因素參雜下,欲使控制測量成果維持良好精度實屬不易,當控制測量成果品質不佳時,易造成相關應用測量成果謬誤。此外現行作業模式中,控制測量網形平差是基於控制點完全無誤的假設前提去平差計算,此方法有誤差累積的潛在問題,控制測量成果品質就會因誤差長久累積而逐漸劣化。本研究以控制點間之相對距離及角度等兩種不受坐標系統或測量方法變動影響之不變觀測量為平差之基礎觀測量,結合已考量各平差參數精度之虛擬序列式平差,建構都市地區控制測量成果之平差解算流程,除可避免誤差累積外,同時具有序列式逐步演化的特性,可將每次測量任務中獲取之高品質觀測量以序列式方式帶入求解,故可使控制測量網系隨時間逐步更新及提升成果品質。在數值實驗部分,除了透過模擬案例進行可行性驗證之外,並以臺北市政府土地開發總隊102年間於士林區專案辦理控制測量成果進行實例分析,經依相關法令規範(如第一章第二節所述)抽驗虛擬序列式平差坐標成果結果,距離抽樣檢測合格比率為93%,角度檢測及坐標抽樣檢測合格比率均為100%,顯示本研究方法確可融入到都市地區控制網系之作業流程,並偵測網系演化過程中之可能問題,藉以確保不同時期之觀測數據品質,並提昇都市地區控制網系之長期穩定性。 In order to manage land more efficiently, the government agencies undergo cadastral surveys to draw cadastral maps for defining the ranges of land rights. The prior task for producing figures and information with high precision to protect people’s rights and interests is to establish sophisticated results of control surveys. Owing to the mixed factors of cadastral surveys across different years, conversion of different coordinate systems, different measure approaches, and frequent public infrastructure constructions, it is not an easy task to maintain the good precision of results of control surveys in urban areas. When the quality of results of control surveys is not good, it causes errors in related results of applied surveys. Besides, in the existing mode of operations, control survey adjustments of networks are based on the assumptions that the control points are fixed and errorless, which causes the potential problems of error accumulation. The consequences of long-term cumulative errors are the gradual deterioration of control surveys.This study combined two invariant observations unaffected by coordinate system or survey methods: the relative distances and angles between surveyed points to serve as the fundamental observations for adjustment, and at the same time, the study integrated the precision of each adjustment parameter in an unified sequential adjustment for considerations. Adjustment computation in results of control surveys of building up urban areas does not only avoid the error accumulation but also shows gradual evolutionary characters. The high-quality measurements derived from each measuring task could be solved with serial approaches. Therefore, the network system can update and elevate the quality of results of control surveys step by step. In numerical experiment, verification of the feasibility could be confirmed by simulation cases, and the results of control survey project conducted by Department of Land Administration, Taipei City Government in Shilin District in 2013 is selected for case analysis to sample and test the results of unified sequential adjustment according to related norms and regulations (described as the first and the second chapters). The qualified ratio of sampling of distance is 93%; testing of angles, and sampling tests of coordinate are all 100%, which indicates that the proposed method could be integrated into the operation procedures of control network in urban areas to detect the potential problems in the evolutionary processes of network-based detection, ensure the qualities of observed data in different time periods and improve the long term stability of control network in urban areas. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/49629 |
DOI: | 10.6342/NTU201602510 |
Fulltext Rights: | 有償授權 |
Appears in Collections: | 土木工程學系 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-105-1.pdf Restricted Access | 2.5 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.