This　paper　puts　forward　one　kind　of　air-fuel　ration　control　method　which　integrates　improved　Elman　neural　network　with　normal　PI　control.　It　constructs　a　model　of　the　single　LPG　electro-plating　engine　and　simulation　platform　for　air-fuel　ration　controlling　with　GT-　Power　software　which　seamless　connect　to　Matlab/Simulink　based　on　JL1P39FMB　single　cylinder　engine　and　develop　the　electronic　fuel-injected　controller　based　on　Intel　MCS-　96.　It　also　constructs　mini　LPG　electro-plating　engine　experiment　system　for　air-fuel　ration　to　the　LPG　Injection　System　and　adopt　the　new　inlet　channel　injection　type　of　duty　ratio　controlling　injector.　Help　the　throttle　percentage-rev-duty　cycle　pulse　spectrum　diagram　after　the　calibration　bench　experiment　for　the　best　duty　cycle　of　mini　electronic　LPG-injected　under　the　steady　working　conditions.　It　predicts　the　air-fuel　ration　signals　of　nontransmission　delay　through　the　Elman　neural　network.　The　normal　PI　controller　which　deals　with　the　predictive　signals　implements　the　concurrent　control　of　air-fuel　ration　under　transient　conditions.　The　bench　test　and　the　simulation　result　indicate　that　the　control　methods　have　the　strong　adaptation,　which　can　make　the　statism　of　air-fuel　ration　under　transient　conditions　on　±5%.　©　2010　ACADEMY　PUBLISHER.