Empirical　mode　decomposition　(EMD)　is　a　self-adaptive　analysis　method　for　nonlinear　and　non-stationary　signals.　It　may　decompose　a　complicated　signal　into　a　collection　of　intrinsic　mode　functions　(IMFs)　based　on　the　local　characteristic　time　scale　of　the　signal.　The　EMD　method　has　attracted　considerable　attention　and　been　widely　applied　to　fault　diagnosis　of　rotating　machinery　recently.　However,　it　cannot　reveal　the　signal　characteristic　information　accurately　because　of　the　problem　of　mode　mixing.　To　alleviate　the　mode　mixing　problem　occurring　in　EMD,　ensemble　empirical　mode　decomposition　(EEMD)　is　presented.　With　EEMD,　the　components　with　truly　physical　meaning　can　be　extracted　from　the　signal.　Utilizing　the　advantage　of　EEMD,　this　paper　proposes　a　new　EEMD-based　method　for　fault　diagnosis　of　rotating　machinery.　First,　a　simulation　signal　is　used　to　test　the　performance　of　the　method　based　on　EEMD.　Then,　the　proposed　method　is　applied　to　rub-impact　fault　diagnosis　of　a　power　generator　and　early　rub-impact　fault　diagnosis　of　a　heavy　oil　catalytic　cracking　machine　set.　Finally,　by　comparing　its　application　results　with　those　of　the　EMD　method,　the　superiority　of　the　proposed　method　based　on　EEMD　is　demonstrated　in　extracting　fault　characteristic　information　of　rotating　machinery.　(C)　2008　Elsevier　Ltd.　All　rights　reserved.