To　understand　the　frequent　bulge　fault　of　hydraulic　operating　mechanism　of　the　breaker　in　a　substation,　oil　pump　failure　is　analyzed,　and　liquid　level　and　gas　sensors　are　used　to　design　an　intelligent　exhaust　device,　which　weakens　the　influence　of　the　environment　and　improves　the　safety　and　reliability　of　the　equipment.　This　device,　with　neural　network　self-learning　height　measurement　algorithm　as　its　core,　processes　the　data　obtained　by　barometric　sensor　and　liquid-level　sensor　by　detecting　the　breaker,　to　get　an　accurate　oil-level　height　and　reach　real-time　control　over　solenoid　valve　exhaust　device　as　well　as　automatic　record　of　system　working　condition,　thus　realizing　the　intelligent　detection　and　automatic　identification.