A new methodology for damage assessment of bridges through instrumentation : application to the Sorraia River Bridge

Abstract

Bridge Management Systems (BMSs) provide an efficient means of manipulating information about a bridge in order that its long-term health is assured and maintenance programmes can be formulated in line with budgetary constraints and funding limitations. However, some weak points remain unsolved and thus limit the possibilities of BMSs. Since accurate material deterioration and damage evolution models are not yet fully developed, one of the weaknesses of a BMS is how to decide when to act and which level of intensity should be used in maintenance/repair/strengthening operations. A high level of uncertainty is associated with the decision of undertaking these operations as the available deterioration models (chloride ingress, carbonation, alkali – silica reaction) are still not completely accurate due to the many variables involved (environmental, material, loading). Structural monitoring is a possible way of dealing with such a problem. Normally, in situ measurements have been used just to update deterioration models included in the damage modules of a BMS. However, in this paper a different approach is presented. Data from the instrumentation are used not to feed the damage models, but are directly used in a new module of the BMS, the so-called monitoring/control module with the objective of detecting degradation when incipient so the decision module of the BMS may respond appropriately. This module should perform damage detection in a fast, effective and accurate way. In order to achieve that, a new methodology, Interval Analysis, and a new algorithm were developed. They have been checked in laboratory tests where the potential for damage detection has been compared to other techniques. They are now being applied for the first time in a prestressed concrete bridge in Portugal. The follow-up of this application will be the basis for the adoption of the new monitoring/control module in a Portuguese BMS.