Cutting quality prediction of a quasi-5-axis abrasive waterjet machine with an adjustable workhead

Forward tilted water jet cutting can eliminate the cutting defects to improve some AWJ (Abrasive Water Jet) cutting performances. The paper studies the cutting quality prediction of a quasi-5-axis AWJ machine with an adjustable tilted workhead. Three prediction models are built based on approaches of discriminant classification, BP network, and PNN. The cutting quality is evaluated by the formalism of classification. After experimental verification and analysis, the result shows that the prediction model of cutting quality based on BP network enjoys the highest prediction accuracy and the prediction error can meet the practical demand.