Prediction of Geometric Dimensions of Deposited Layer Produced Using Laser-Arc Hybrid Additive Manufacturing

Area of Science:

• Materials Science and Engineering
• Manufacturing Technology
• Computational Modeling

Background:

• Laser-arc hybrid additive manufacturing (LAHAM) offers industrial potential but faces challenges in dimensional accuracy.
• Precise control over deposited layer dimensions (width and height) is critical for forming precision in LAHAM.
• Existing methods struggle to accurately predict and manage geometric variations in LAHAM processes.

Purpose of the Study:

• To investigate the influence of process parameters on the width and height of deposited layers in LAHAM.
• To develop and validate a predictive model for accurate forecasting of layer dimensions.
• To enhance dimensional control and improve the overall forming precision of LAHAM.

Main Methods:

• Experimental analysis of laser power, arc current, and scanning speed effects on layer dimensions.
• Application of the Taguchi method for quantitative analysis of parameter significance.
• Development of a predictive model using extreme gradient boosting (XGBoost) optimized with particle swarm optimization (PSO).

Main Results:

• Layer width positively correlates with laser power and arc current; negatively with scanning speed.
• Layer height negatively correlates with laser power and scanning speed; positively with arc current.
• The PSO-XGBoost model demonstrated superior accuracy and lower error metrics (MRE, MSE, R²) compared to PSO-SVR and baseline XGBoost.

Conclusions:

• Arc current is the most influential parameter on deposited layer dimensions, followed by scanning speed.
• The developed PSO-XGBoost model effectively captures complex nonlinear relationships for precise dimensional prediction in LAHAM.
• This study offers valuable insights for optimizing LAHAM process parameters to achieve desired geometric accuracy.