Dynamic analysis and system identification of a synchronous belt mechanism driven by PMSM

In this paper, Newton's law and Euler equation is employed to derive dynamic formulation of a SPE workstation driven by PMSM. The dynamic formulations can be expressed by some independent variables. In order to verify this dynamic formulation is correct, we reduce the flexible model to the rigid one under some assumptions, and adopt the genetic algorithm (GA) to identify all the parameters of the mechanism and PMSM simultaneously. It turned out that the GA can identify model parameters which are difficult to measure in practical, such as inductance, stator resistance, motor torque constant, damping coefficient of the PMSM and timing belt. In order to verify the identification result is feasible, the angular displacements of the timing belt mechanism driven by PMSM was obtained in the numerical simulations and experimental results. From their comparison, it suggest that the dynamic model identified sufficient for a input-output model within a modelling accuracy of -0.3-0.3rad, such the identified model presents better matching with experimental results of the system.