LPV model identification using blended linear models with given weightings

Nonlinear process identification is studied. In model identification, a linear parameter varying (LPV) model is used and it consists of weighted local linear models. In this work, predetermined weighting functions are used and the LPV model identification becomes a linear identification problem with multiple weighted input data sets. So simplicity is its strength. The developed method is especially suitable for batch processes for which only transition tests are feasible and no working point tests are permitted. It is also suitable for continuous nonlinear process identification. Identifiability and stability of the LPV model are discussed. Simulation studies will be used to verify the effectiveness of the method.