The　motor　is　one　of　the　most　commonly　used　equipment　in　the　industry.　It　is　necessary　to　ensure　the　reliability　of　the　motor,　and　identify　the　type　of　motor　fault　in　time　to　ensure　the　normal　operation　of　the　motor　and　reduce　the　loss.　In　this　paper,　a　turn-to-turn　short　circuit　of　motor　stator　and　unbalance　power　supply　fault　diagnosis　system　based　on　Deep　Auto-Encoder　and　Soft-max　Classifier　is　proposed.　The　influence　of　neural　network　parameters　on　the　training　process　and　the　choice　of　parameters　are　given.　The　proposed　fault　diagnosis　system　can　map　the　motor　state　to　a　2-dimension　vector,　corresponding　to　different　area　of　a　plane　to　identify　different　fault　type.　Finally,　the　proposed　system　is　verified　by　experiment　on　a　motor　in　laboratory.　The　conclusion　shows　the　ability　to　identify　the　fault　type　of　motor　at　the　continuous　state　that　the　accuracy　is　above　99.5%,　when　only　the　data　from　motor　at　discrete　state　point　are　used　in　training,　which　makes　the　system　extensible　and　promising.　©　2018　IEEE.