Abstract: Aiming at the problem of high false alarm rate of the condition monitoring system caused by complex and changeable operating conditions of the wind turbine main transmission system,a method for monitoring the operational state of the wind turbine main transmission system based on working condition recognition is proposed. Firstly,in order to solve the problem of unclear operating conditions due to complex and variable operational state of wind turbines,the SCADA feature parameters are selected by the maximal information coefficient（MIC）,and the operating conditions of the main transmission system are classified by the k-means clustering algorithm. Then the CMS feature parameters are extracted,and the relative weights of CMS feature parameters are calculated by analytic hierarchy process（AHP）,and the evaluation index and quantification algorithm of the main transmission system operating condition is proposed. Finally,the threshold value is determined using the kernel density estimation（KDE） method,and the abnormality monitoring of the main transmission system is realized according to the relationship between the threshold value and the operating condition index. The proposed method is applied to a wind farm,and the experimental results show that the monitoring results are consistent with the actual situation.