Design And Optimization Of Low Ripple Synchronous Reluctance Machine

With the continuous development of modern motor industry and the urgent need for free rare earth permanent magnet motors,specialists and scholars at home and abroad have been attracted by synchronous reluctance motors(SynRM)because of its high efficiency,simple structure,low cost and convenient control performance.However,SynRM’s application is restricted by the disadvantages of SynRM,such as high torque ripple,low average torque and power factor in new energy vehicles and other strategic emerging industries.SynRM design and optimization are becoming increasingly difficult due to the complex multi-layer magnetic barrier rotor structure.In addition,the deterministic multi-objective optimization methods do not include the uncertainties in the course of SynRM rotor’s processing and assembling into multi-objective optimization process.Therefore,the design of SynRM rotor structure is studied in the paper.The deterministic multi-objective optimization based on sequential subspace and the robust multi-objective optimization based on Six Sigma criteria and Latin hypercube sampling are researched to aiming at improving the performance and robustness of SynRM in an all-round way.The research contents have both theoretical innovation significance and engineering application value.Based on the 3kW and 4-pole SynRM,the research on the low torque ripple SynRM topology is carried out.Firstly,the 36-slot and three-layer magnetic barrier SynRM structure is selected by comparison different slot numbers and magnetic barrier layers.To improve average torque and power factor,the central magnetic bridge of magnetic barrier is removed and the thickness of magnetic rib is reduced.To further reduce the torque ripple at same time improve the average torque and power factor,the V type step-skewed pole is adopted in the axial direction and the left and right asymmetric magnetic barrier structures is used in radial direction respectively.The finite element model was established in ANSYS MAXWELL to verify the effectiveness of the scheme.Finally,the linear magnetic barrier end was changed to the flat magnetic barrier end to improve the SynRM mechanical performance and mechanical checking is completed.Aiming at SynRM’s complex multi-layer magnetic barrier structure and overcoming the SynRM’s shortcomings of average torque,torque ripple and power factor,the related research on deterministic multi-objective optimization is carried out.The partial parameters of each layer of the magnetic barrier are set to the same value to simplify the rotor structure and reduce the number of design variables considering motor performance.Then,the weight of the design variables on the objective function is obtained by the Taguchi method.The key optimization variables are selected according to the results.The whole optimization space is divided into three sub-spaces,significantly important sub-space,important sub-space and non-important sub-space.Finally,the deterministic multi-objective optimization is implemented for each subspace based on differential evolution algorithm and Pareto evaluation in the order.In view of the influence of parameter uncertainty on motor performance,the related research on robust multi-objective optimization of SynRM is carried out.Firstly,the parameter uncertainty of the SynRM rotor structure is simulated by the error of processing and the normal distribution law.The parameter range is narrowed to avoid the optimization variable exceeding the range of values due to the uncertainty.Then,the sample points are obtained from the normal distribution using Latin hypercube sampling.Through the statistical formula,the target performance and its robustness metric the mean μ and the standard deviation σ are obtained.Then the deterministic multi-objective optimization based on sequence subspace is used to obtain the Pareto front.The target performance range is set according to the optimization results.The solutions satisfying the performance range are selected in all of population.Finally the robustness evaluation is carried out and compared to obtain the optimal performance and the robust motor design.