Global optimization by coupled local minimizers and its application to FE model updating

Abstract

Coupled local minimizers (CLM) is a new method applicable to global optimization of functions with multiple local minima. In CLM a cooperative search mechanism is set up using a population of local optimizers, which are coupled during the search process by synchronization constraints. CLM is characterised by a relative fast convergence since the local optimizers are gradient-based. The combination of both, the coupled parallel strategy and the fast convergence, offers an efficient global optimization algorithm. In the paper the CLM method is described and is illustrated with a test function. Due to the simultaneous and coupled search of a whole population of optimizers, CLM is able to find the global minimum of the test function. Next, CLM is successfully applied to FE model updating using experimental modal data. In an example the damage pattern of a reinforced concrete beam is identified.