PROBABILISTIC SAFETY ASSESSMENT OF CIVIL ENGINEERING STRUCTURES WITH AN EMPHASIS ON BRIDGES

摘要

Bridges are examples of important elements of our infrastructure. Therefore, nowadays one of the most important tasks of building authorities is to manage and maintain the safety and usability of the existing bridge stock. Normally, all bridges in Germany are inspected every three or six years respectively. If deterioration becomes too advanced, they are repaired. The problem is that in times of shortened budgets not every necessary maintenance or rehabilitation measure can be carried out. To ensure the safety of a deteriorated structure, structural health monitoring (SHM) can be used. The exceedance of threshold values during the SHM process indicates a further increase of damage or deterioration. In such cases, the usage of the structure should be stopped immediately before serious situations can occur. The highest safety could be achieved if the complete structure is monitored. In most cases, this is not possible, because such a monitoring strategy would be expensive. In the Collaborative Research Center 477 (CRC 477) funded by the German research council (DFG) at the Braunschweig University of Technology, methods for the optimization of SHM are investigated. In project field A1 of the CRC a framework for reliability-based system assessment based on SHM is developed. The methodology combines recognized procedures of system and reliability theory for the optimization of monitoring measures. After a thorough anamnesis of a structure, typical weak points and failure paths of the structure are identified. The knowledge gained about the structural system is used to formulate a probabilistic model of the system, which is evaluated in reliability analyses using the first and second order reliability methods (FORM/SORM). Within the framework, the reliability analyses are carried out in two steps. In the first step, the actual state of the structure is evaluated, which is usually done when enough data from the monitoring process is available and significant changes in the structural behavior are detected. In a second step, a prognosis calculation is carried out to estimate the development of the reliability level of the structure in future. With the help of the results of the reliability analysis, the weak points and failure paths can be weighted. Additionally the most important parameters of the structure are identified. This information can be used to concentrate the monitoring measures only on the most significant parts of the structure. The time intervals between the safety evaluations depend on the results of the prognosis calculations. The developed methods of the framework are integrated into the computer code PROBILAS (PRObabilistic Building Inspection and Life ASsessment), which is one additional research topic of project field A1 of CRC 477. This article presents and explains the elements of the framework and shows its application using a substitute bridge structure constructed in the CRC 477. A substitute structure represents a section of a real structure, which can be seen as a large-scale model. The actions on these substitute structures can be influenced and controlled. In this special case, the structure is subjected to controlled deterioration as well. Usually, concrete structures are quite sensitive for depassivation of the reinforcement and the following corrosion. Therefore, sodium chloride is applied systematically in certain areas of the substitute structure. After a period of time this will lead to a corrosion of the reinforcement and eventually the pre-stressing elements. With the occurring chloride ingress the sensor equipment and evaluation procedures can be tested and validated. The focus in the example at the end of this article is the probabilistic assessment of the substitute structure with special regard to the chloride ingress in the concrete of the structure.