An Ensemble Random Forest Model to Predict Bead Geometry in GMAW Process

Abstract

A prevalent method for rapid prototyping of metallic parts is gas metal arc welding (GMAW). As the input parameters impose a highly nonlinear impact on the weld bead geometry, precise estimation of the geometry is a complex problem. Therefore, in this study, a novel combination of the most powerful machine learning algorithms is selected to overcome the complexity of the problem and also reach an acceptable degree of precision. To this end, the hybrid combination of the support vector machine (SVM) and relevance vector machine (RVM) is developed based on the random forest (RF) ensemble learning approach. The models are established based on a global database of welding geometry, and the corresponding process parameters obtained are based on a set of experiments. Performance evaluation between RVM, SVM, and the proposed model was performed based on the coefficient of determination ([Formula: see text]) and the ratio of root means square error (RMSE) to the maximum measured outputs ([Formula: see text]/[Formula: see text]). The RF-based RVM-SVM model obtained 0.9725 and 0.8850 for [Formula: see text] and 0.0257 and 0.0447 for [Formula: see text]/[Formula: see text] in predicting the height and width of the bead, respectively. The result clearly showed the effectiveness of the proposed model in predicting the GMAW trend.

Heat input prediction during dissimilar welding of steel using different machine learning model

“…A significant number studied were conducted on dissimilar welding for better understanding the underlying concept to produce the sound weld in terms of better mechanical properties of the joint.A large number of experiment needs to be conducted by varying the different input parameter in order to produce sound welds which is certainly a time consuming process and also it requires significant human effort.To resolve these issue, researcher round the world are using different machine learning models to predict the output characteristics based on the given input variables which effect these characteristics [5,6,7,8].A team of researcher [10] developed a hybrid model combining support vector machine (SVM) and relevance vector machine (RVM) to predict the bead geometry during gas metal arc welding of metallic parts. The performance measured in terms of root mean square error (RMSE) is observed to be 0.0257 and 0.0447 in predicting the height and width of the bead respectively.…”