The　failure　of　operating　mechanism　occupies　the　absolute　majority　of　the　high　voltage　circuit　breaker　failure.　Identification　and　prediction　of　operating　mechanism　of　the　failure　of　high　voltage　circuit　has　a　great　effect　in　breaker　maintenance　and　overhaul.　In　this　paper,　a　mechanical　status　identification　method　based　on　covariance　matrix　is　used　to　get　mechanical　characteristic　signals　from　the　contact　travel　curve.　The　theory　of　Principle　Component　Analysis　(PCA)　is　a　projection　from　high-dimensional　data　to　low-dimensional　data.　Then　it　extracts　the　main　characteristic　of　the　sample　to　identify　other　data.　To　use　PCA,　a　large　amount　of　data　of　the　breaker　under　fault　condition　is　needed.　But　to　simulate　the　fault　condition　of　the　breaker,　some　experiments　are　not　easy　to　do　to　get　the　data　we　want　and　some　experiments　are　very　harmful　to　the　breaker.　In　this　paper,　based　on　multi-body　dynamics　theory　and　method,　a　high　voltage　circuit　model　has　been　developed　to　carry　out　the　high　voltage　circuit　breaker　dynamics　simulation.　The　simulation　of　the　opening　and　closing　process　of　the　high-voltage　circuit　breaker　under　varieties　of　typical　fault　conditions　is　carried　out　to　get　speed　curve　and　travel　curve.　The　mechanical　characteristic　signals　of　typical　failure　types　are　extracted　from　a　large　number　of　simulation　data.　The　typical　failure　types　of　high　voltage　circuit　breaker　can　be　identified　and　predicted　in　the　work　through　the　mechanical　characteristic　signals.