.png){kind=icon}
.png){kind=icon}
Return
Machine learning-based prediction of machining parameters in nanoclay-filled GRPC
.png){kind=icon}
PRE
AI
U
N
DOI：10.1080/10426914.2026.2698482.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Nanoclay-filled glass fiber-reinforced polymer composites (GFRPCs) have attracted substantial attention in the field aerospace and automobile sectors due to excellent mechanical properties, low weight, and better durability. However, because of their heterogeneous structure, machining is difficult and requires optimal process parameters to produce a high-quality product. This study reports a machine learning-based approach to predict machining performance. Among the ML models used, the study revealed that XGBoost showed the best performance for surface prediction (Ra) prediction with R² = 0.8194 and MAPE = 4.66%, whereas Polynomial Regression (2°) had perfect performance for MRR prediction with Test R² = 0.99 and MAPE = 0.00%. Feed rate and nanoclay concentration are found to be key factors for Ra, whereas feed rate and depth of cut are found to be key factors for MRR. Bayesian Ridge and SVR are found ineffective for MRR, and all ensemble methods overfit the data for Ra.
Keywords:
Surface roughness
MRR
machine learning
XGBoost
Polynomial Regression
Journal
M
IF:
4.7
Papers:
4.6K
Citations:
9.3K
Organization
Cited Papers