The 2023 IEEE International Conference on Predictive Control of Electrical Drives and Power Electronics (PRECEDE 2023)

In order to take the nonlinear characteristics of friction into account, Stribeck model is adopted for modeling. However, it is an obstacle to identify all the parameters owing to the nonlinear characteristic. On the other hand, the flux linkage mismatch brings about the errors in identification. Aiming at the above problems, a flux-linkage-independence equivalent model is built and a two-step method is proposed to identify all equivalent mechanical parameters. Eventually, online controller tuning based on equivalent mass identification and friction compensation are implemented. The experiments are carried out on an ironless permanent magnet synchronous linear motor. The results show that the identified parameter converges accurately and fast. Moreover, the proposed method performs better than the other two methods in position control.

Controller tuning, friction compensation, mechanical parameter identification, permanent magnet synchronous linear motor, recursive least square, Stribeck model.