Identification of hysteretic systems using the differential evolution algorithm

A widely used model in the field of hysteretic or memory-dependent vibratio ns is that of Bone and Wen. Different parameter values extend its use to va rious areas of mechanical vibrations. As a consequence an identification me thod is required to identify the parameter values relevant to its applicati on. Its structure, however, includes internal states and non-linear terms. This rules out the conventional identification methods, such as least squar es and maximum likelihood because they require derivative calculations of t he prediction error with respect to the parameters. In this paper are prese nted some results for Bouc-Wen model identification, using simulated noise- free data, simulated noisy data and experimental data obtained from a nucle ar power plant. The method used to achieve this is the differential evoluti on algorithm. Differential evolution (DE) is an optimization method develop ed to perform direct search in a continuous parameter space without requiri ng any derivative estimation. (C) 2001 Academic Press.