Characterization and damage identification of acoustic emission signal in tensile process of the material of high-speed train gearbox shell

Abstract

In operating conditions, real time non-destructive testing (NDT) is needed for the identification of tensile damage process of high-speed train gearbox shell. This paper focuses on the application of an acoustic emission (AE) method to study tensile damage. First, tensile tests with AE monitoring were employed to collect AE signals and tensile damage data. Second, feature extraction was performed to obtain feature vectors to characterize the process of tensile damage. Then, Support Vector Machines (SVM) and Weighted Support Vector Machines (WSVM) were used to identify different tensile damage stages of the material of high-speed train gearbox shell.