This　article　aims　to　use　data-driven　control　to　achieve　adaptive　dynamic　optimization　of　the　electromagnetic　switch　during　the　pull-in　process　and　holding　stage.　First　of　all,　through　the　use　of　the　characteristics　of　the　repeated　operation　of　the　electromagnetic　switch,　the　control　principle　of　the　data　drive　for　the　switch　pull-in　and　hold　process　is　clarified.　Subsequently,　a　data-driven　model　of　electromagnetic　switch　dynamic　optimization　was　constructed.　The　proposed　RBR　control　process　does　not　depend　on　the　specific　parameters　of　the　electromagnetic　switch.　It　can　perform　feedback　evaluation　and　analysis　on　the　historical　action　information　of　the　switch,　so　that　the　advantages　of　feedforward　and　feedback　are　complementary　and　versatile.　Finally,　an　RBR　data-driven　intelligent　decision-making　scheme　based　on　current　scanning　is　proposed,　which　is　a　closed-loop　reference　sequence　that　optimizes　the　excitation　current　and　its　action　time　by　rolling.　Relevant　simulations　and　experiments　verify　the　effectiveness　of　the　control　strategy,　which　can　effectively　reduce　the　contact　bounce　of　the　contactor　and　the　dispersion　of　the　action　time.　And　changes　the　limitations　of　the　existing　mathematical　model-based　control　scheme,　which　always　guide　the　contactor　online　time-varying　operation　by　obtaining　the　optimal　control　parameters　offline.　©　2021,　Beijing　Oriental　Sun　Cult.　Comm.　CO　Ltd.