City Research Online

Functional and stochastic models for geometrical detection of spatial deformation in engineering: A practical approach

Setan, H. B. (1995). Functional and stochastic models for geometrical detection of spatial deformation in engineering: A practical approach. (Unpublished Doctoral thesis, City, University of London)

Abstract

The objective of this study is to formulate a simple and practical but rigorous two-step analysis procedure for the geometrical detection of spatial deformation using geodetic methods. A thorough and critical study of theory and current practice of deformation monitoring has been undertaken, and a practical scheme has been developed for 3-D least squares estimation (LSE) and one-stage detection procedure (i.e. stability determination and localization of spatial deformation) via two-epoch analysis.

In LSE, a simple datum definition via minimum constraints with fixed coordinates has been adopted; a strategy for rank defect analysis of normal equations by simplified eigenvalue decomposition (EVD) has been developed; an optimised computational procedure for S-transformations has been formulated; a mathematical model for additional parameters and pseudo observables (distance differences and ratios) has been extended and established for 3-D application; a procedure for handling of algebraically correlated pseudo observations via observation de-correlation has been established; a procedure for robustified LSE for multiple gross errors detection has been formulated and its effects has been derived; a simple method of variance component estimation (VCE) has been extended; and the use of global and local tests and reliability analyses in LSE has been presented.

In deformation detection, a strategy for determination of common stations between epochs via S-transformations and partitioning has been developed; a flexible one-stage computational procedure for geometrical detection of spatial deformation by iterative congruency testing and S-transformations has been established; the robust method for deformation detection has been modified to allow one-stage computation; and general S-transformations equations have been applied in all cases.

This developed strategy has been implemented in five computer programs (ESTIMATE, COMPS, COMON, DETECT and ROBUST). The developed programs can be executed either on an IBM based personal computer (PC) or under the UNIX environment. Links between these programs and two of the Engineering Surveying Research Centre’s (ESRC) programs (GAP and DCRE) have been established. The programs have been successfully applied and evaluated using simulated and real data. Five real photogrammetric monitoring schemes undertaken by the ESRC, with up to 169 stations, were analysed for detecting the significance of spatial deformation between epochs. The results obtained confirmed the suitability of the strategy in practical applications.

Further refinement to the developed programs are suggested to make them more user friendly. Further possible research activities include a combined or integrated approach for deformation analysis and real time deformation monitoring.

Publication Type: Thesis (Doctoral)
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Departments: School of Science & Technology > Engineering > Civil Engineering
School of Science & Technology > School of Science & Technology Doctoral Theses
Doctoral Theses
[img]
Preview
Text - Accepted Version
Download (10MB) | Preview

Export

Downloads

Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login