Advanced machine learning framework for hole quality prediction and optimization in polycarbonate AWJD

This study presents a methodology-oriented framework for predicting and optimizing hole quality in abrasive waterjet drilling (AWJD) of polycarbonate materials. Utilizing experimental datasets generated from a full factorial design, the influence of key process parameters—specifically water pressure, standoff distance, and traverse rate—on performance metrics including hole diameter error and drilling rate is systematically analyzed. Statistical validation via analysis of variance identifies water pressure and standoff distance as the primary factors governing machining quality. To model the highly non-linear relationships inherent in the AWJD process, four machine learning algorithms—namely XGBoost, Decision Tree, Random Forest, and AdaBoost Regressor—are developed and evaluated. Among these, the Random Forest model serves as a highly reliable

surrogate paradigm due to its superior predictive accuracy. This surrogate model is subsequently integrated with metaheuristic optimization techniques, including the Grey Wolf Optimizer, Whale Optimization Algorithm, and Arithmetic Optimization Algorithm, to identify optimal parameter configurations that minimize diameter error while maximizing the drilling rate. The framework provides structured guidance for advanced manufacturing process control, offering a robust computational approach to multi-objective optimization in polymer machining.

Message: Hi FE, I need help with “Advanced machine learning framework for hole quality prediction and optimization in polycarbonate AWJD” in “Design & Manufacturing / Advanced Manufacturing”