Abstract: An online methodology of abnormality detection of electrohydraulic servo valve’s working condition was studied. A new feature extraction method based on the valve’s dynamic characteristics was proposed. One class support vector machine (SVM) was adopted to identify the working condition of the valve and the SVM parameters were optimized by crossvalidity estimation learning means. The results indicate that the proposed feature extraction method can effectively extract the fault features of the valve and its abnormal condition can be more effectively identified by taking advantage of oneclass support vector machine’s generalization ability than by the neural network, which lays the foundation for online faults diagnosis of electrohydraulic servo valve.