Metal　oxide　arrester　(MOA)　is　the　key　equipment　to　protect　the　power　system　from　voltage　shock,　and　its　normal　operation　is　of　great　significance　to　ensure　the　stable　operation　of　the　power　system.　However,　there　are　a　large　number　of　false　alarms　and　missing　alarms　in　the　current　online　monitoring　of　MOA.　Therefore,　in　view　of　the　problems　existing　in　the　current　rules　of　online　monitoring　of　MOA,　firstly,　the　temperature　and　humidity　shielding　is　realized　by　adopting　the　temperature　and　humidity　interference　suppression　method　based　on　weighted　nonlinear　surface　modeling;　Secondly,　the　inverse　cumulative　distribution　function　is　used　to　calculate　the　alarm　threshold　according　to　the　historical　data　of　lightning　arrester,　and　the　alarm　rules　for　early　warning　of　lightning　arrester　defects　are　constructed;　finally,　the　effectiveness　of　the　early　warning　rule　is　demonstrated　by　an　example　analysis.　©　2022　IEEE.