Optimization of fixture locating layout design using comprehensive optimized machine learning

Abstract Fixtures are commonly employed in production as work holding devices that keep the workpiece immobilized while machined. The workpiece’s deformation, which affects machining precision, is greatly influenced by positioning of fixture elements around workpiece. By locators and clamps appropriately, deformation might be decreased. Therefore, it required to model fixture–workpiece system’s complicated behavioral relationship. In this study, long short-term memory (LSTM), multilayer perception (MLP), adaptive neuro-fuzzy inference system (ANFIS) three machine-learning approaches connection between locator clamp positions maximum throughout end milling. hyperparameters developed ANFIS, MLP, LSTM chosen using evolutionary algorithms, including genetic algorithm (GA), particle swarm optimization (PSO), butterfly (BOA), grey wolf (GWO), (WOA). Among methods, MLP optimized BOA (BOA-MLP) reached highest accuracy among all models predicting response surface. had a lower computational load than final element calculating surface during process. At step, prementioned five algorithms were implemented BOA-MLP extract optimal parameters decrease deflection machining. proposed method was modeled MATLAB. outcomes showed mentioned efficient enough compared with previous method, such methodology point view 0.0441 μm deflection.