Datasets:
Tasks:
Tabular Regression
Languages:
English
Size:
1K<n<10K
ArXiv:
Tags:
computational-fluid-dynamics
surrogate-modeling
proper-orthogonal-decomposition
electronics-cooling
thermal-management
reduced-order-model
License:
| """ | |
| 02 — Plot the first 6 POD spatial mode shapes on the yz-mid plane. | |
| Refits PCA with 6 components on the full dataset (the canonical model | |
| in 01 only retains N_MODES=4, but for visualization we want a couple | |
| extra modes to illustrate the diminishing structure). | |
| """ | |
| import numpy as np | |
| import matplotlib.pyplot as plt | |
| from sklearn.decomposition import PCA | |
| from _utils import load_data, PLOTS_DIR | |
| N_PLOT = 6 | |
| def main(): | |
| _, fields, coords, _ = load_data() | |
| pca = PCA(n_components=N_PLOT).fit(fields) | |
| var = pca.explained_variance_ratio_ | |
| y, z = coords[:, 1], coords[:, 2] | |
| fig, axes = plt.subplots(3, 2, figsize=(8, 7)) | |
| for i, ax in enumerate(axes.flatten()): | |
| mode = pca.components_[i] | |
| vmax = np.abs(mode).max() | |
| ax.scatter(z, y, c=mode, cmap="RdBu_r", s=1.5, | |
| vmin=-vmax, vmax=vmax, rasterized=True) | |
| ax.invert_xaxis() | |
| ax.set_xticks([]); ax.set_yticks([]) | |
| ax.set_aspect("equal") | |
| ax.set_title(f"PC{i+1} | Var: {var[i]*100:.2f}%", fontsize=14) | |
| plt.subplots_adjust(hspace=0.05, wspace=0.05) | |
| PLOTS_DIR.mkdir(parents=True, exist_ok=True) | |
| out = PLOTS_DIR / "pod_modes.png" | |
| plt.savefig(out, dpi=300, bbox_inches="tight") | |
| print(f"Saved {out}") | |
| plt.close() | |
| if __name__ == "__main__": | |
| main() | |