MRAS Based Parameter Identification and Integration on Deadbeat-Direct Torque and Flux Control

Advanced machine control techniques, such as MPC and DB-DTFC, relies heavily on accurate machine parameters and inaccurate parameters can degrade control performance. Since machine parameters vary due to multiple factors, it is necessary to embed real-time parameter identification method into these control techniques. This paper is dedicated to exploring parameter identification method integrated on DB-DTFC. Firstly, the parameter sensitivity of DB-DTFC is studied. Then, making use of a flux observer combining the current and voltage model, a MRAS based parameter estimator is constructed to obtain magnetizing inductance and rotor time constant of the machine. Finally, experimental results show the validity of our method.