With　the　increase　in　electricity　demand,　the　rising　maintenance　costs　of　substation　projects　have　put　power　grid　companies　under　huge　market　pressure.　Traditional　cost　methods　cannot　meet　the　needs　of　modern　markets,　so　the　construction　of　intelligent　cost　calculation　models　is　the　key　to　realize　the　modernization　and　refinement　of　capital　management　and　control　of　power　grid　enterprises.　In　this　paper,　the　cost　composition　of　the　primary　equipment　maintenance　of　the　substation　is　analyzed,　and　the　engineering　parameters,　equipment　parameters　and　process　parameters　are　determined　to　be　the　key　influencing　factors　of　the　total　project　cost　by　analyzing　the　historical　engineering　settlement　data.　On　this　basis,　the　deep　belief　network　was　used　to　establish　the　measurement　model　for　maintenance　engineering　cost　prediction.　Experiments　are　carried　out　on　the　actual　engineering　data　of　Guangdong　Province,　and　the　prediction　results　show　that　the　proposed　calculation　method　and　model　have　good　accuracy　and　reliability.　　©　2025　IEEE.