Decoupling control based on neural network inverse for bearingless synchronous reluctance motor

In the paper, the principle of a bearingless synchronous reluctance motor (BSRM) is explained and its mathematical model is established, then decoupling control method based on neural network inverse system is put forward, and inverse system of the bearingless synchronous reluctance motor is constructed by adopting neural network inverse system to achieve the decoupling of electromagnetic torque and suspension radial force. The gained inverse models are in series before the original system to decouple a complex nonlinear multivariable system into 4 relatively independent single input single output (SISO) pseudo-linear subsystems. The close loop linear controllers and stator flux observer are designed, and the digital simulation experiment has been carried out by using Matlab software. The theory research and simulation experiment results have validated that decoupling control in the transient case can be achieved successfully adopting the inverse system strategy, good performance of dynamic and static state of system can be also obtained, and validity of the proposed control method is demonstrated.