Three-vector Predictive Current Control Of Five-arm Double Permanent Magnet Motor

As a good fault-tolerant structure,the five-leg dual-permanent magnet synchronous motor system is widely used in many fields such as aerospace and rail transportation.In the multi-motor control system,it is necessary to consider optimization of multiple control objectives in order to ensure that the system is in an optimal operating state.Model predictive control,as an online optimization control algorithm,directly generates driving signals and acts on voltage source inverters when processing nonlinear system problems.Model predictive control combines modulation and control in one.The principle is simple and it is easy to achieve multi-conditional constraints and multi-objective optimization of the system.Since there exists a common leg in the five-leg dual-motor system,it is necessary to ensure that the switching state of the common leg meets the two motors’ requirements when controlling the system.The traditional research method is to substitute the switching state of the five-leg inverter into the value function as a whole,then select the optimal vector to act on the five-leg inverter.The traditional model predictive current control only acts on one vector with fixed amplitude and direction in each control cycle,and the range of vector regulation is limited,which reduces the control accuracy of the system.In order to improve the control accuracy of the system,a three-vector predictive current control strategy for the five-leg double motor system is proposed in this thesis from the perspective of expanding the range of vector regulation.In each control cycle,two effective vectors and one zero vector are applied to make the predicted value of d-q axis current equal to the expected value at the next moment.The amplitude and direction of the desired voltage vector synthesized in each control cycle can be adjusted,which enlarges the range of vector regulation.Firstly,the effective vectors of the two motors are grouped according to the switch state of the common leg corresponding to the two effective vectors,and the optimal vector combination in each group is screened out by the respective value function.And then the optimal vector combination of the system is screened out by the system value function,while the five-leg voltage vector is synthesized and applied to the five-leg inverter according to the set rules.The proposed three-vector control strategy is equivalent to controlling two three-leg inverters independently drive a dual-motor system.The traditional predictive current control strategy and the proposed three-vector predictive current control strategy are simulated in Matlab/Simulink,and the experiments are completed on the experimental platform.The simulation and experimental results show that the proposed three-vector predictive current control strategy effectively reduce the d-q axis current fluctuation and improve the system control accuracy while having good dynamic effects.