Surface Roughness Prediction of Titanium Alloy during Abrasive Belt Grinding Based on an Improved Radial Basis...

Area of Science:

• Materials Science
• Manufacturing Engineering
• Computational Intelligence

Background:

• Titanium alloys offer superior strength and corrosion resistance, making them widely used.
• Grinding titanium alloys presents significant challenges due to their unique material properties.
• Accurate surface roughness prediction is essential for optimizing machining parameters in real-time.

Purpose of the Study:

• To develop an advanced theoretical model for predicting the surface roughness of ground titanium alloy.
• To enhance the accuracy and reliability of surface roughness forecasting in titanium alloy machining.

Main Methods:

• An improved radial basis function neural network (RBF) model was developed.
• The model integrated particle swarm optimization (PSO) with the grey wolf optimization (GWO) method (GWO-PSO-RBF).
• The GWO-PSO-RBF model was employed to forecast surface roughness during titanium alloy grinding.

Main Results:

• The developed GWO-PSO-RBF model demonstrated superior performance over classical prediction models.
• The model achieved a high model-fitting coefficient of determination (R²) of 0.919.
• The enhanced neural network provided more accurate prediction parameters for surface roughness.

Conclusions:

• The GWO-PSO-RBF model offers a significant advancement in predicting titanium alloy grinding surface roughness.
• This improved model can facilitate real-time adjustments to machining parameters for better outcomes.
• The study highlights the potential of hybrid optimization algorithms in precision manufacturing applications.