From fault detection to one-class severity discrimination of 3D printers with one-class support vector machine

The lack of faulty condition data reduces the feasibility supervised learning for fault detection or severity discrimination in new manufacturing technologies. To deal with this issue, one-class arises building binary discriminative models using only healthy data. However, these have not been extrapolated to discrimination. This paper proposes extend OCSVM, which is typically used detection, 3D printer First, a set features extracted from normal signals. An optimized OCSVM model obtained by tuning kernel and hyperparameters. resulting are evaluated proposed performance evaluation approach. Experimental comparisons belt-based faults printers show that distance hyperplane has information discriminate level, its use feasible. hyperparameter optimization technique improves compared some other methods.