Research On Model Predictive Control For A Seven-Phase Induction Motor

Compared with three-phase induction motor, the number of stator winding in multi-phase induction motor is increased. With it, the electromagnetic torque ripple is reduced, and the fault tolerance of the speed regulating system is improved while the power requirement of each phase stator winding is decreased. However, because multiphase motor speed control system is multivariable and strong coupling, the variables that we need to control and the control complexity will increase with the add of the number of stator windings. Model Predictive Control (MPC) has a huge advantage in the treatment of multivariable constrained control problems, especially in multiphase motor speed control system, thus, it has attracted a lot of attention of scholars both domestic and overseas about how to apply it to the speed control system of multiphase motor.Thisthesis is based on seven-phase induction motor. Firstly, a mathematical model is established under the natural coordinate system. The 7 dimensional mathematical model in the natural coordinate system is transformed into the 2 dimensional decoupling space by the decoupling matrix which we have been derived, it reduces the control complexity of the seven phase motor.Secondly, a novel PWM, seven-phase UVM (Unified Voltage Module), is put forward and applied into the current control of seven-phase voltage source inverter. Based on the idea of vector control of three-phase induction motor, the method of vector control of seven-phase induction motor base on UVM is derived. The simulation results show that this control strategy has good dynamic and static performance, but it needs many PI adjustors and makes the debugging of the system too complex.Besides, the method of finite control set model predictive control (FCS-MPC) is introduced into the current control of seven-phase voltage source inverter. And then the influence of some factors on the control performance is studied, including the objective function, the number of space vector and the sample time. For the problem that model predictive control computation in unit time is too large, combined with the sector where the fundamental space required in, the improved method is proposed to reduce calculation and improve the control performance. Based on the improved method, the model predictive control of seven-phase induction motor is put forward. The simulation results show that the model predictive control,without complex Debug, has better performance than the vector control which is based on UVM.Finally, the experiment platform of seven-phase induction motor speed control system is built, and then the relevant experiments are executed. The experimental results and simulation results are consistent, verifying the correctness of the theory put forward by this paper.