Parameter Estimation for Non-Linear Systems

Abstract

Due to the many structural possibilities of non-linear relations between the input and output of dynamic systems, one cannot expect to be able to identify many types of non-linear system with only a few model classes. However, for certain types of nonlinear systems, models can be formulated that match well with the requirements on the model structure of known identification methods. In this sense, some model structures and suitable parameter identification methods will be covered in the following. First, dynamic systems with continuously differentiable non-linearities will be discussed, then dynamic systems with non-continuously differentiable non-linearities, such as friction and dead zone will be treated.