Abstract:Whirling is one kind of abnormal vibration in deep hole drilling. During the drilling process, it may enlarge the roundness error, and even damage the tool and hole wall. Thus, it is important to monitor the drilling condition and identify the whirling timely. To be specific, the whirling shouldbe suppressedtimelyand thequality ofdeep hole machining should be improved. In thisstudy, anonlinewhirlingdetectionmethodin deep holedrillingisproposedbased onvibrationsignal.Firstly,thevibration signal isdecomposedby empirical wavelet transform, and the high multiple frequency components of spindle rotation are extracted. Secondly, the energy ratio between the extracted component and the original signal is calculated. Finally, the energy ratio is viewed as the detection index to identify the tool condition. The proposed method is evaluated with BTA deephole drilling tests. Experimental results show that the proposed method can effectively identify the whirling which lead to the roundness error that is larger than 0035 mm during the deep hole drilling.