Linear switched system modelling and analysis for the servo system

Aiming at resolving the parameter uncertainty of the servo system under multi-working conditions, a flatness-based switched linear system model is proposed. Firstly, a fifth-order linear switched discrete-time system with uncertain parameters is established. The flatness properties of switched system is described and a de Bruijn's graph-based framework is introduced for graphical description of the switched system. In addition, non-linear bode diagrams are obtained via sweeping signals with multiple amplitudes. The discrete equations of the electromechanical servo system are established by the bode plots. The unknown parameters of the subsystems in switched model are identified by the simulated annealing algorithm, and compared with the least squares method to verify the accuracy of the identified parameters. Finally, the simulation model and experimental platform are constructed to verify the effectiveness of the proposed method. The switched model is compared with the classical transfer function model and single subsystems. The results indicate that the resolution of parameter uncertainty in servo systems is effective. The accuracy of the established switched model is about 99.5%. The accuracy of the switched linear system model of the electromechanical servo system is verified and the proposed modelling method is validated. A new switched linear system model is proposed to better model the electromechanical servo systems. The generalised theory of switching linear discrete-time systems with parameter uncertainty is developed, and the flatness between past and future inputs is characterised. And, the validity of the accuracy for the switched model and the modelling approach presented are demonstrated in the experiments.image