|
Organizers |
Optimisation of structural monitoring using reliability-based system assessment
by
D. Hosser
Institut für Baustoffe, Massivbau und Brandschutz, TU-Braunschweig
Coauthors: M. Dehne, R. Schnetgöke
For reasons of economy, structural monitoring has to be concentrated on the weak spots critical for the structural behaviour and the corresponding decisive uncertainties. In order to accomplish this, methods for the identification of such weak points and uncertainties are used for the definition of optimal monitoring measures as well as assessment and decision criteria. These methods are based on recognized procedures of reliability and system theory. In order to make the application possible to building monitoring engineers without special training in reliability theory, the methods were summarized in the knowledge-based system PROBILAS (PRObabilistic Building Inspection and Life ASsessment). This computer code consists of a data base module, a computation module and a statistics and updating module, which are linked by a user-friendly graphical user interface and serve for the optimisation of the building assessment cycle. An essential component of the assessment is the acquisition and illustration of the building as a system and its integration into the data base and the computation module of PROBILAS. Since the logical models of real buildings, e.g. bridges, needed as elements of the system reliability computation, can be very complex, methods are developed to seize and integrate the possible failure mechanisms. In this article, firstly the building assessment cycle with the knowledge-based system PROBILAS is illustrated. The continuous revaluation of the system and the focussing of both the stochastic and the physical models on the failure-relevant parts of the system, limit states and parameters are characteristics of this cycle. A main focus of this article is on the methods of system integration. Some steps of the system generation run more or less automatically, e.g. the creation of response surfaces for the limit-state functions of system components. In other domains the monitoring engineer is consciously involved into the process while PROBILAS offers the necessary assessment and decision criteria. The different methods are elucidated and afterwards illustrated by the example of a bridge construction.
Date received: December 21, 2003
Copyright © 2003 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # canb-72.