Table 9 Summary of BPANN architecture, training settings, and performance metrics.
Parameter | BPANN Model Details |
|---|---|
Model Type | Back-Propagation Artificial Neural Network (BPANN) |
Input Features | Geometry Type (One-Hot), Wall Thickness (mm), Load (kN) |
Output Targets | Displacement (mm), Strain |
Total Data Samples | 27 |
Training Samples | 18 |
Testing Samples | 9 |
Activation Function (Hidden Layer) | ReLU |
Activation Function (Output Layer) | Linear |
Number of Hidden Layers | 1 |
Number of Neurons | 12 |
Loss Function | Mean Squared Error (MSE) |
Optimizer | Adam |
Learning Rate | 0.01 |
Epochs | 1000 |
Batch Size | 4 |
Train-Test Split Ratio | 2:1 (18 training, 9 testing) |
