State space identification of bilinear continuous-time canonical systems via batch scheme Hartley modulating functions approach

摘要

This paper deals with the state space identification of bilinear continuous-time dynamic systems represented in canonical forms based on the evaluation of input and noise corrupted output batch data. Bilinear continuous-time canonical forms are an important class of nonlinear systems which may be found in a wide variety of applications and they are suitable for solving analysis and synthesis problems of dynamic systems. The batch scheme Hartley modulating functions (1942) identification method starts from a priori knowledge of the bilinear system structure with unknown parameters. The unknown system parameters can then be estimated by shifting a batch data one step forward at each sampling time applying a frequency weighted least squares algorithm. Simulation studies are considered to examine the performance of the algorithm.