In　order　to　solve　the　problem　that　the　performance　parameters　of　permanent　magnet　actuators　are　difficult　to　be　identified　and　by　computers　and　the　multi-objective　optimization　problem,　this　paper　proposes　a　multi-objective　optimization　method　based　on　the　surrogate　model.　Firstly,　a　surrogate　model　based　on　orthogonal　least　squares　radial　basis　functions　(OLS-RBF)　is　proposed　to　solve　the　problem　of　identifying　the　performance　parameters　of　permanent　magnet　actuators.　Then,　for　the　multi-objective　optimization　problem,　a　novel　elimination　mechanism　is　proposed　for　NSGA-III　to　take　the　place　of　the　original　selection　mechanism,　which　improves　the　diversity　of　the　Pareto　solution　sets　and　reduces　the　running　time　of　the　algorithm.　And　combine　it　with　the　fitted　surrogate　model　for　multi-objective　optimization.　Finally,　taking　the　electromagnetic　needle　selector　with　permanent　magnets　as　an　example,　the　feasibility　of　the　method　and　the　accuracy　of　the　improved　NSGA-III　algorithm　are　verified　by　simulation,　which　lays　the　foundation　for　the　general　optimization　design　scheme　of　permanent　magnet　actuators.　©　2024,　Beijing　Paike　Culture　Commu.　Co.,　Ltd.