Stochastic Linearization and Large Deviations
Stochastic equivalent linearization is the most popular approximation method for the dynamic of a non-linear system under random excitation. A complete presentation of this method can be found in . Despite the fact it was introduced 40 years ago, the first justification was proposed by F.Kozin  in 1987. Another approach was recently introduced by the author in collaboration with L. Wu , based on the use of a large deviation principle. The goal of this contribution is to present this approach to the stochastic dynamic engineering public.
KeywordsProbability Measure Markov Process Relative Entropy Polish Space Empirical Process
Unable to display preview. Download preview PDF.
- 1.Alaoui Ismaili M.(1995) PHD thesis, Université Blaise PascalGoogle Scholar
- 2.Bernard P., Wu L.(1995) Linéarisation d’un oscillateur excité par un bruit blanc: un point de vue entropique, Report of the Laboratoire de Mathématiques Appliquées, Université Biaise PascalGoogle Scholar
- 3.Kozin F.(1988) The Method of Statistical Linearization for Non-linear Stochastic Vibrations, in Nonlinear Stochastic Dynamic Engineering Systems, F.Ziegler, G.I.Schueller Editors, Springer VerlagGoogle Scholar
- 5.S.R.S.Varadhan S.R.S.(1984) Large Deviations and Applications, SIAM Publications 46.Google Scholar