HydroModPy/tools/view_mesh_grid_3d.py at dev · HydroModPy/HydroModPy · GitHub

History

347 lines (304 loc) · 13.1 KB

Raw

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

#!/usr/bin/env python

"""Standalone 3D viewer for a HydroModPy catchment mesh bundle.

Reads ONLY the self-contained mesh bundle CSVs (no hydromodpy import) plus the

lake geometry, extrudes every planar triangle into a prism (z_bottom -> z_top),

exports the 3D grid to VTU (open in ParaView / pyvista) and renders PNG views:

the full prismatic aquifer grid and a zoom on the lakes' unstructured cells.

The bundle is produced next to the mesh under

``<project>/mesh/mesh_catchment_bundle/`` and is documented in its README.md.

Usage:

python tools/view_mesh_grid_3d.py \

--bundle examples/projects/19_cheze_reservoir/mesh/mesh_catchment_bundle \

--lakes examples/data/lake_geometry/lakes_cheze_preretenue.gpkg \

--out /tmp/grid3d --zexag 25

"""

from __future__ import annotations

import argparse

from pathlib import Path

import geopandas as gpd

import numpy as np

import pandas as pd

from shapely.geometry import Point

from shapely.prepared import prep

def parse_args() -> argparse.Namespace:

p = argparse.ArgumentParser(description=__doc__)

p.add_argument("--bundle", required=True, type=Path, help="mesh_catchment_bundle dir")

p.add_argument("--lakes", type=Path, default=None, help="lake geometry gpkg (optional)")

p.add_argument("--out", type=Path, required=True, help="output dir for VTU + PNG")

p.add_argument("--zexag", type=float, default=25.0, help="vertical exaggeration for the render")

p.add_argument(

"--keep-islands",

action="store_true",

help="keep enclosed non-lake cells (real islands); default fills them so the lake is contiguous",

)

return p.parse_args()

def load_bundle(bundle: Path):

nodes = pd.read_csv(bundle / "nodes.csv").sort_values("node_id").reset_index(drop=True)

cells = pd.read_csv(bundle / "cells.csv")

if not (nodes.node_id.to_numpy() == np.arange(len(nodes))).all():

raise SystemExit("node_id is not a dense 0..N-1 range; remap needed.")

tri = cells[["n0", "n1", "n2"]].to_numpy(dtype=np.int64)

return nodes, cells, tri

def classify_lakes(cells: pd.DataFrame, lakes_path: Path | None):

"""Return (lake_id per cell, {id: name}). 0 = aquifer, 1..K = each lake."""

lake_id = np.zeros(len(cells), dtype=np.int64)

names = {0: "aquifer"}

if lakes_path is None or not lakes_path.exists():

return lake_id, names

lk = gpd.read_file(lakes_path).to_crs("EPSG:2154")

centroids = np.c_[cells.centroid_x.to_numpy(), cells.centroid_y.to_numpy()]

for k, (_, row) in enumerate(lk.iterrows(), start=1):

prepared = prep(row.geometry)

inside = np.fromiter(

(prepared.covers(Point(px, py)) for px, py in centroids),

dtype=bool,

count=len(centroids),

)

lake_id[inside] = k

names[k] = str(row.get("lake_id", f"lake{k}"))

return lake_id, names

def cell_adjacency(tri: np.ndarray) -> dict:

"""Edge-neighbour lists: cells sharing 2 nodes are adjacent."""

from collections import defaultdict

edge2cell = defaultdict(list)

for ci, (a, b, c) in enumerate(tri):

for e in ((a, b), (b, c), (a, c)):

edge2cell[tuple(sorted(e))].append(ci)

nbr = defaultdict(list)

for shared in edge2cell.values():

if len(shared) == 2:

nbr[shared[0]].append(shared[1])

nbr[shared[1]].append(shared[0])

return nbr

def fill_enclosed_holes(lake_id: np.ndarray, tri: np.ndarray) -> tuple[np.ndarray, int]:

"""Reclassify non-lake cells fully surrounded by one lake into that lake.

Removes the small classification holes / sub-cell islands so a lake footprint

is contiguous. Iterated to fill multi-cell pockets. A cell touching the

aquifer (a shoreline cell) or two different lakes is never filled. Real

islands can be preserved by not calling this (``--keep-islands``).

"""

nbr = cell_adjacency(tri)

lake_id = lake_id.copy()

n_filled = 0

while True:

changed = 0

for ci in np.where(lake_id == 0)[0]:

neighbours = nbr.get(int(ci), [])

if neighbours and all(lake_id[n] > 0 for n in neighbours):

vals = {int(lake_id[n]) for n in neighbours}

if len(vals) == 1:

lake_id[ci] = vals.pop()

changed += 1

n_filled += changed

if changed == 0:

return lake_id, n_filled

def build_wedge_grid(nodes: pd.DataFrame, tri: np.ndarray, cells: pd.DataFrame, lake_id):

"""Extrude each triangle into a VTK wedge (true elevations, no exaggeration)."""

import pyvista as pv

n = len(nodes)

x, y = nodes.x.to_numpy(), nodes.y.to_numpy()

zt, zb = nodes.z_top.to_numpy(), nodes.z_bottom.to_numpy()

pts = np.empty((2 * n, 3), dtype=float)

pts[:n, 0], pts[:n, 1], pts[:n, 2] = x, y, zb # bottom sheet

pts[n:, 0], pts[n:, 1], pts[n:, 2] = x, y, zt # top sheet

n_c = len(tri)

conn = np.empty((n_c, 7), dtype=np.int64)

conn[:, 0] = 6

conn[:, 1:4] = tri

conn[:, 4:7] = tri + n

celltypes = np.full(n_c, pv.CellType.WEDGE, dtype=np.uint8)

grid = pv.UnstructuredGrid(conn.ravel(), celltypes, pts)

grid.cell_data["z_top"] = cells.z_top_mean.to_numpy()

grid.cell_data["thickness"] = (cells.z_top_mean - cells.z_bottom_mean).to_numpy()

grid.cell_data["lake_id"] = lake_id

return grid

def render_pyvista(grid, lake_id, names, out: Path, zexag: float) -> bool:

import pyvista as pv

pv.OFF_SCREEN = True

try:

pv.start_xvfb()

except Exception:

pass

lake_colors = ["royalblue", "deepskyblue", "turquoise", "teal"]

try:

# Full 3D prismatic grid, aquifer by elevation + lakes highlighted.

pl = pv.Plotter(off_screen=True, window_size=(1500, 950))

aqu = grid.extract_cells(np.where(lake_id == 0)[0])

pl.add_mesh(

aqu,

scalars="z_top",

cmap="terrain",

show_edges=False,

scalar_bar_args={"title": "z_top [m]"},

)

for k in sorted(set(lake_id) - {0}):

pl.add_mesh(

grid.extract_cells(np.where(lake_id == k)[0]),

color=lake_colors[(k - 1) % len(lake_colors)],

show_edges=True,

)

pl.add_text(f"Cheze catchment - 3D prismatic grid (z x{zexag:g})", font_size=11)

pl.set_scale(zscale=zexag)

pl.show_grid()

pl.view_isometric()

pl.screenshot(str(out / "grid3d_full.png"))

pl.close()

# Zoom on the lakes' unstructured cells.

if set(lake_id) - {0}:

pl = pv.Plotter(off_screen=True, window_size=(1500, 950))

for k in sorted(set(lake_id) - {0}):

sub = grid.extract_cells(np.where(lake_id == k)[0])

pl.add_mesh(

sub,

color=lake_colors[(k - 1) % len(lake_colors)],

show_edges=True,

label=f"{names[k]} ({int((lake_id == k).sum())} cells)",

)

pl.add_legend()

pl.add_text("Lakes - unstructured triangular cells", font_size=11)

pl.set_scale(zscale=zexag)

pl.show_grid()

pl.view_isometric()

pl.screenshot(str(out / "grid3d_lakes.png"))

pl.close()

return True

except Exception as exc: # noqa: BLE001 - rendering backend may lack GL

print(f"[pyvista render skipped: {exc!r}] VTU export is still written.")

return False

def render_matplotlib(nodes, tri, cells, lake_id, names, out: Path, zexag: float) -> None:

"""Guaranteed static PNGs via matplotlib (top surface + lake cells)."""

import matplotlib

matplotlib.use("Agg")

import matplotlib.pyplot as plt

from matplotlib.cm import ScalarMappable

from mpl_toolkits.mplot3d.art3d import Poly3DCollection

x, y = nodes.x.to_numpy(), nodes.y.to_numpy()

zt = nodes.z_top.to_numpy()

verts_top = np.stack([x[tri], y[tri], zt[tri] * zexag], axis=-1) # (ncell,3,3)

ztc = cells.z_top_mean.to_numpy()

norm = plt.Normalize(ztc.min(), ztc.max())

facecol = plt.cm.terrain(norm(ztc))

is_lake = lake_id > 0

facecol[is_lake] = plt.cm.tab10(0) # blue for lakes

fig = plt.figure(figsize=(13, 8))

ax = fig.add_subplot(111, projection="3d")

coll = Poly3DCollection(verts_top, facecolors=facecol, edgecolors="none", linewidths=0)

ax.add_collection3d(coll)

ax.set_xlim(x.min(), x.max())

ax.set_ylim(y.min(), y.max())

ax.set_zlim((zt.min()) * zexag, (zt.max()) * zexag)

ax.set_box_aspect((np.ptp(x), np.ptp(y), np.ptp(zt) * zexag))

ax.set_title(f"Cheze catchment - top surface, {len(tri)} triangular cells (z x{zexag:g})")

ax.set_xlabel("x [m]")

ax.set_ylabel("y [m]")

ax.view_init(elev=35, azim=-120)

sm = ScalarMappable(norm=norm, cmap="terrain")

fig.colorbar(sm, ax=ax, shrink=0.5, label="z_top [m]")

fig.tight_layout()

fig.savefig(out / "grid3d_full_mpl.png", dpi=130)

plt.close(fig)

# Lake cells only, top view.

if is_lake.any():

fig, ax = plt.subplots(figsize=(11, 7))

colors = {1: "royalblue", 2: "deepskyblue", 3: "turquoise"}

for k in sorted(set(lake_id) - {0}):

m = lake_id == k

polys = [np.c_[x[t], y[t]] for t in tri[m]]

from matplotlib.collections import PolyCollection

ax.add_collection(

PolyCollection(

polys,

facecolors=colors.get(k, "teal"),

edgecolors="k",

linewidths=0.3,

label=f"{names[k]} ({int(m.sum())} cells)",

)

lm = is_lake

ax.set_xlim(x[tri[lm]].min(), x[tri[lm]].max())

ax.set_ylim(y[tri[lm]].min(), y[tri[lm]].max())

ax.set_aspect("equal")

ax.set_title("Lakes - unstructured triangular mesh (top view)")

ax.set_xlabel("x [m]")

ax.set_ylabel("y [m]")

ax.legend(loc="upper right")

fig.tight_layout()

fig.savefig(out / "grid3d_lakes_mpl.png", dpi=140)

plt.close(fig)

def render_top2d(nodes, tri, cells, lake_id, names, out: Path) -> None:

"""Plain 2D top view of the full unstructured triangular mesh."""

import matplotlib

matplotlib.use("Agg")

import matplotlib.pyplot as plt

from matplotlib.cm import ScalarMappable

from matplotlib.collections import PolyCollection

x, y = nodes.x.to_numpy(), nodes.y.to_numpy()

polys = [np.c_[x[t], y[t]] for t in tri]

ztc = cells.z_top_mean.to_numpy()

norm = plt.Normalize(ztc.min(), ztc.max())

fig, ax = plt.subplots(figsize=(13, 7))

# All cells: terrain fill + thin edges so the triangulation is visible.

ax.add_collection(

PolyCollection(

polys,

array=ztc,

cmap="terrain",

norm=norm,

edgecolors="0.35",

linewidths=0.12,

)

# Lakes overlaid with a solid colour + crisper edges.

lake_colors = {1: "royalblue", 2: "deepskyblue", 3: "turquoise"}

for k in sorted(set(lake_id) - {0}):

m = lake_id == k

ax.add_collection(

PolyCollection(

[polys[i] for i in np.where(m)[0]],

facecolors=lake_colors.get(k, "teal"),

edgecolors="k",

linewidths=0.25,

label=f"{names[k]} ({int(m.sum())} cells)",

)

ax.set_xlim(x.min(), x.max())

ax.set_ylim(y.min(), y.max())

ax.set_aspect("equal")

ax.set_xlabel("x [m]")

ax.set_ylabel("y [m]")

ax.set_title(f"Unstructured triangular mesh - top view ({len(tri)} cells)")

sm = ScalarMappable(norm=norm, cmap="terrain")

fig.colorbar(sm, ax=ax, shrink=0.7, label="z_top [m]")

if set(lake_id) - {0}:

ax.legend(loc="upper left", fontsize=8)

fig.tight_layout()

fig.savefig(out / "grid2d_top.png", dpi=160)

plt.close(fig)

def main() -> None:

args = parse_args()

args.out.mkdir(parents=True, exist_ok=True)

nodes, cells, tri = load_bundle(args.bundle)

lake_id, names = classify_lakes(cells, args.lakes)

if not args.keep_islands and (lake_id > 0).any():

lake_id, n_filled = fill_enclosed_holes(lake_id, tri)

if n_filled:

print(

f"filled {n_filled} enclosed lake hole cell(s); use --keep-islands to preserve them"

)

print(f"mesh: {len(nodes)} nodes, {len(cells)} triangular cells")

for k in sorted(set(lake_id)):

print(f" {names[k]:16s}: {int((lake_id == k).sum())} cells")

try:

grid = build_wedge_grid(nodes, tri, cells, lake_id)

grid.save(args.out / "cheze_grid3d.vtu")

if (lake_id > 0).any():

grid.extract_cells(np.where(lake_id > 0)[0]).save(args.out / "cheze_lakes3d.vtu")

print(f"VTU written to {args.out} (open in ParaView / pyvista)")

render_pyvista(grid, lake_id, names, args.out, args.zexag)

except Exception as exc: # noqa: BLE001

print(f"[pyvista path failed: {exc!r}] falling back to matplotlib only.")

render_matplotlib(nodes, tri, cells, lake_id, names, args.out, args.zexag)

render_top2d(nodes, tri, cells, lake_id, names, args.out)

print(f"PNG views written to {args.out}")

if __name__ == "__main__":

main()

HydroModPy/tools/view_mesh_grid_3d.py at dev · HydroModPy/HydroModPy · GitHub

Uh oh!

Expand file tree

Breadcrumbs

view_mesh_grid_3d.py

Latest commit

History

Breadcrumbs

view_mesh_grid_3d.py

File metadata and controls