| As the basic power equipment in the industrial field,the promotion and application of high-efficiency and energy-saving induction motor is an important method to improve energy utilization rate and realize energy saving and emission reduction.And in the production and use of motor,there are inevitable uncertainties and uncontrollable factors,such as the manufacturing process of the manufacturer.Due to these uncertainties,even if the motor produced in the same batch,there is the probability that the motor performance of the final product will fail below the standard requirements.Therefore,carrying out the research of the design and multi-objective robust optimization of ultrahigh efficiency induction motor is urgent and having both important theoretical research significance and engineering application value.Based on a 4 poles,22 k W induction motor,the stator winding is improved firstly,the design method including the reasonable determination of the cross-sectional area of the wire,the arrangement of the winding and the number of conductors in each slot.The efficiency is improved by mainly reducing the copper loss of the induction motor through these method.And the radial skewed rotor is adopt in rotor side which could improve the efficiency by reducing the harmonic content of the air gap and improve the torque performance and power factor of induction motor at the same time.The validity of the electromagnetic structure is proved later by the finite element modeling and analysis of the induction.Then set efficiency,power factor and the torque ripple as the optimization objectives,based on the finite element analysis model,the Taguchi method,differential evaluation algorithm and Pareto evaluation is combined to complete the sequential multiobjective optimization of ultra-high efficiency induction,and the Pareto solution set will be obtained and the optimal design will be obtained finally.Considering the influence of various uncertain factors on the performance of induction motor in the manufacturing process and the problem of the design cycle based on the finite element analysis.Firstly,the optimal Pareto solution set of the deterministic optimization result is taken as the sample space and the subspace combination surrogate model is established by combining the radial basic function model and the polynomial response surface model according to the subspace division in the process of the deterministic optimization.Then based on the combined surrogate model and the Latin hypercube sampling,the perturbation is added to all design combinations on the optimal Pareto solution set that meet the performance requirements.And the robustness of these design combinations are evaluated based on the design for 6σ.Finally,the research goal of robust design of ultra-high efficiency induction motor is achieved.The synchronization robust optimization design method and the distributed robust optimization design method proposed in this paper are compared finally.In order to make a more fair comparison,the robustness evaluation criteria,the method of adding perturbation to the optimization parameters,the type of the surrogate model and the optimization algorithm are set the same.By comparing the characteristic of the two methods,including the design flow,the accuracy of the optimization results and the calculation time,the application situation is determined,which provides design reference and guidance for the multi-objective robust optimization of ultra-high efficiency induction motor. |
