Concurrent multifactor optimisation techniques for model updating of long-span bridges

This article aims at developing an innovative technique of concurrent multi-factor optimisation for updating the multi-scale model of civil infrastructure. A multi-scale model is established for the purpose of concurrently analysing the global response of the structure and non-linear damage of local details in order to meet the needs of evaluation on structural state as well as structural deteriorating. Traditional model updating methods cannot be applied directly to updating such type of models. The technique of concurrent multi-factor optimisation technique (CMFOT) developed in the article can update the multi-scale model by optimising simultaneously the different factors to affect on structural parameters. The key issues of the developed concurrent multifactor optimisation technique are discussed in details, and implementation of the model updating with CMFOT is presented by a case study of a welded steel truss with the available tested data. The multi-scale model of the truss is updated by using the proposed technique and other model updating methods in order to compare their difference. The proposed technique is also applied to updating the concurrent multi-scale model of a long-span suspension bridge, the Runyang Suspension Bridge (RYSB). The results show that CMFOT method is the improvement of traditional model updating methods. The main advantage of CMFOT is that the factors to be considered in CMFOT include not only those sensitive to the updated objective but also other factors sensitive to other objectives to be not updated currently. The computational efficiency of model updating by the CMFOT is higher than others when the correct sequence of multi factors updating is unknown. It is necessary to have sufficient measuring data from the field tests with careful arrangement of the measuring points for the purpose of the model updating with CMFOT for large civil infrastructure like RYSB.