Fault detection on the 3-D printed objective surface by using the SVM algorithm

The concept of 3D-printed items has gained traction in the age of Industry 4.0, proving helpful in terms of money and time spent. Assemble the products like this are physically in stages according to computerised CAD data. However, faults still exist in 3D printed objects owing to the diversity in characteristics and structure, which degrades printed quality. Must uncover these mistakes at each development stage of the product. This study Using an ensemble of machine learning algorithms and pre-trained models, details a procedure for detecting anomalies at each layer. For defect detection, the recommended combination is learned in advance offline and then applied in real-time online. This research compares several pre-trained models using SVM, machine learning methods for monitoring and defect detection in Fused Deposition Modeling in an experimental setting (FDM). According to the findings, the best accuracy was achieved by combining Alexnet with the SVM algorithm. As a result of its minimal experimental and computational costs, apply the suggested fault detection technique for real-time defect detection.