Analytical Calculation Of Magnetic Field And Cogging Torque Optimization Of Partitioned Stator Flux Reversal Permanent Magnet Machine Based On Subdomain Model

Partitioned Stator Flux Reversal Permanent Magnet Machine(PS-FRPM)is a new type of Stator-PM Machines,which is evolved from traditional Flux Reversal Permanent Magnet Machine(FRPM).The Permanent-Magnets and windings of PM-FRPM are respectively placed on the inner and outer stator,which solves the place confliction between permanent magnets and armature windings of single stator FRPM,and further increases the power density.In addition,the separation of armature windings and permanent magnets is more convenient for temperature management,thereby reducing the risk of permanent magnet demagnetization,and has potential application prospects in low-speed direct drive fields such as wind power generation and wave power generation.However,its special structure of two-layer air gap and two-layer slotting will cause the cogging torque of the inner and outer air gaps to be superimposed,resulting in the total cogging torque to be too large and affecting the steady-state operation of the machine.Therefore,it is necessary to suppress the cogging torque in such kind of machines.This article takes 12/10 pole PS-FRPM as the research object.Firstly,the research background and significance of the PS-FRPM are introduced,combined with domestic and foreign literature,the development history of the FRPM is reviewed,and the current research status of the PS-FRPM is summarized.Then,based on the principle of general air-gap field modulation theory for electrical machines,the working principle of the PS-FRPM is explained.At present,the finite element method is commonly used in the analysis and research of the PS-FRPM,but its complex structure will increase the complexity of meshing,further increase the magnetic field solution time,and the efficiency is low.Therefore,this paper uses the subdomain model method to solve the magnetic field of the PS-FRPM,then compares the solution results and the solution speed of the subdomain model method and the finite element method to verify the accuracy and rapidity of the deduced subdomain model.Then,based on the subdomain model method,the influence of the structural parameters of the PS-FRPM on the cogging torque and the no-load back EMF is analyzed by means of parameter sweep,and the mean contradiction that affect the cogging torque is found.On this basis,the genetic algorithm is introduced into the cogging torque optimization design of this machine,a multi-objective optimization function is constructed based on the per-unit weight coefficient on the cogging torque and the no-load back EMF performance,and in order to reduce the optimization design cycle,the structural parameter that has a greater impact on the cogging torque is selected and optimized.The optimized machine greatly suppresses the cogging torque without sacrificing the no-load back-EMF performance.Finally,a stepped modulation rotor is proposed to replace the traditional modulation rotor to further suppress cogging torque.Taking a 3-layer stepped modulation rotor as an example,the influence of the pole arc coefficient of each stepped block on the cogging torque and the no-load back EMF performance is analyzed,and the genetic algorithm is also used for multi-objective optimization.The results show that the optimized stepped modulation rotor has a better suppression effect on the cogging torque and does not greatly sacrifice the no-load back EMF performance.