Optimization of a clamping concept based on machine learning

Abstract Fixtures are an important element of the manufacturing system, as they ensure productive and accurate machining differently shaped workpieces. Regarding fixture design or layout elements, a high static dynamic stiffness fixtures is therefore required to defined position orientation workpieces under process loads, e.g. cutting forces. Nowadays, with increase in computing performance development new algorithms, machine learning (ML) offers appropriate possibility use regression methods for creating realistic, rapid reliable equivalent ML models instead simulations based on finite method (FEM). This research work introduces novel that allows optimization clamping concepts by means ML, order reduce errors obtain increased accuracy. paper describes preparation dataset training models, systematic selection most promising algorithm relevant criteria, implementation chosen Extreme Gradient Boosting (XGBoost) other comparable analysis their results, validation selected concept.