ashiq24 commited on
Commit
74c72c1
·
1 Parent(s): 045f29c

added dataloader

Browse files
README.md CHANGED
@@ -6,4 +6,55 @@ tags:
6
  - physics
7
  - Fluid-Solid-Interaction
8
  - Fluid-Dynamics
9
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
  - physics
7
  - Fluid-Solid-Interaction
8
  - Fluid-Dynamics
9
+ ---
10
+
11
+ ## Dataset Description: Fluid-Solid Interaction Simulations
12
+
13
+ For each time step \( t \), the simulation records the following variables:
14
+
15
+ ### **Velocity**
16
+ \[
17
+ \mathbf{v}_t = \begin{bmatrix} v_{x,t} \\ v_{y,t} \end{bmatrix}
18
+ \]
19
+ where \( v_{x,t} \) and \( v_{y,t} \) are the velocity components in the \( x \) and \( y \) directions, respectively.
20
+
21
+ ### **Pressure**
22
+ \[
23
+ P_t
24
+ \]
25
+ which represents the pressure field at time \( t \).
26
+
27
+ ### **Displacement**
28
+ \[
29
+ \mathbf{d}_t = \begin{bmatrix} d_{x,t} \\ d_{y,t} \end{bmatrix}
30
+ \]
31
+ where \( d_{x,t} \) and \( d_{y,t} \) denote the displacement in the \( x \) and \( y \) directions, respectively.
32
+
33
+ ---
34
+
35
+ ## **Mesh Representation**
36
+ The initial mesh is given by:
37
+ \[
38
+ \mathbf{M}_0 = \begin{bmatrix} x_1 & y_1 \\ x_2 & y_2 \\ \vdots & \vdots \\ x_N & y_N \end{bmatrix} \in \mathbb{R}^{N \times 2}
39
+ \]
40
+ where each row \( (x_i, y_i) \) represents a mesh point in 2D space.
41
+
42
+ Since the mesh is time-dependent, the mesh at time \( t \) is updated based on displacement:
43
+
44
+ \[
45
+ \mathbf{M}_t = \mathbf{M}_0 + \mathbf{d}_t
46
+ \]
47
+
48
+ where
49
+ \[
50
+ \mathbf{d}_t = \begin{bmatrix} d_{x,t,1} & d_{y,t,1} \\ d_{x,t,2} & d_{y,t,2} \\ \vdots & \vdots \\ d_{x,t,N} & d_{y,t,N} \end{bmatrix} \in \mathbb{R}^{N \times 2}
51
+ \]
52
+ is the displacement field at time \( t \).
53
+
54
+ Thus, the updated coordinates of the mesh points at time \( t \) are:
55
+
56
+ \[
57
+ (x_i^t, y_i^t) = (x_i^0 + d_{x,t,i}, y_i^0 + d_{y,t,i}) \quad \forall i = 1, \dots, N
58
+ \]
59
+
60
+ This formulation describes how the mesh deforms over time due to displacement.
data_vis.ipynb CHANGED
The diff for this file is too large to render. See raw diff
 
fsi_animation.gif → fsi_animation_dx.gif RENAMED
File without changes
fsi_animation_pressue.gif ADDED

Git LFS Details

  • SHA256: 032353545450018b52169ad093f2a0897dc107461233d12c812db3b66dae67b1
  • Pointer size: 133 Bytes
  • Size of remote file: 40.3 MB
fsi_reader.py CHANGED
@@ -8,14 +8,38 @@ import h5py
8
  class FsiDataReader():
9
  def __init__(self,
10
  location,
11
- mu=None,
12
  in_lets_x1=None,
13
  in_lets_x2=None,):
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  self.location = location
15
  self._x1 = ['-4.0', '-2.0', '0.0', '2.0', '4.0', '6.0']
16
  self._x2 = ['-4.0', '-2.0', '0', '2.0', '4.0', '6.0']
17
  self._mu = ['0.1', '0.01', '0.5', '5', '1.0', '10.0']
18
- # keeping vx,xy, P, dx,dy
19
  self.varable_idices = [0, 1, 3, 4, 5]
20
 
21
  if mu is not None:
@@ -33,6 +57,7 @@ class FsiDataReader():
33
 
34
  # assert _mu = 0.5 should not be mixed with other mu values
35
  assert not('0.5' in self._mu and len(self._mu) > 1), "mu=0.5 should not be mixed with other mu values"
 
36
 
37
 
38
  def load_mesh(self, location):
@@ -138,10 +163,10 @@ class FsiDataReader():
138
  for x2 in self._x2:
139
  try:
140
  if mu == 0.5:
141
- mu_data = self.get_data_txt(mu, x1, x2))
142
  else:
143
- mu_data = data.append(self.get_data(mu, x1, x2))
144
- mu_data_t0 = mu_data[:1,:,:]
145
  mu_data_t1 = mu_data[1:,:,:]
146
  data_t0.append(mu_data_t0)
147
  data_t1.append(mu_data_t1)
 
8
  class FsiDataReader():
9
  def __init__(self,
10
  location,
11
+ mu,
12
  in_lets_x1=None,
13
  in_lets_x2=None,):
14
+ '''
15
+ Data set of fluid solid interaction simulations.
16
+ At each time step t, the simulataion records 5 variables:
17
+ velocity_t = [vx_t, vy_t]: velocity in x and y direction,
18
+ P_t: pressure,
19
+ displacment_t = [dx_t, dy_t]: displacement in x and y direction.
20
+
21
+ The inital mesh is loaded as self.input_mesh. The mesh is a 2D mesh with 2 columns. The first column is the x coordinate and the second column is the y coordinate.
22
+
23
+ The mesh is time dependent i.e., the mesh changes with time.
24
+ The mesh at time t is given by mesh_t = self.input_mesh + displacement_t.
25
+
26
+ Parameters
27
+ ----------
28
+ location : str
29
+ path to the directory containing the data
30
+ mu : list, optional
31
+ list mu vlues. The siumulations corresponding to the mu values will be loaded. The values should be one of ['0.1', '0.01', '0.5', '5', '1.0', '10.0'] and slould exactly match the string values given here. The mu='0.5' should not be loaded separately.
32
+ in_lets_x1, : list, optional
33
+ list of x1 parameter controlling the inlet boundary condition of the simulation. The values should be one of ['-4.0', '-2.0', '0.0', '2.0', '4.0', '6.0'] and slould exactly match the string values given here. default is None, which loads all the values.
34
+ in_lets_x2 : list, optional
35
+ list of x2 parameter controlling the inlet boundary condition of the simulation. The values should be one of ['-4.0', '-2.0', '0.0', '2.0', '4.0', '6.0'] and slould exactly match the string values given here. default is None, which loads all the values.
36
+
37
+ '''
38
  self.location = location
39
  self._x1 = ['-4.0', '-2.0', '0.0', '2.0', '4.0', '6.0']
40
  self._x2 = ['-4.0', '-2.0', '0', '2.0', '4.0', '6.0']
41
  self._mu = ['0.1', '0.01', '0.5', '5', '1.0', '10.0']
42
+ # keeping vx, xy, P, dx,dy
43
  self.varable_idices = [0, 1, 3, 4, 5]
44
 
45
  if mu is not None:
 
57
 
58
  # assert _mu = 0.5 should not be mixed with other mu values
59
  assert not('0.5' in self._mu and len(self._mu) > 1), "mu=0.5 should not be mixed with other mu values"
60
+ self.load_mesh(location)
61
 
62
 
63
  def load_mesh(self, location):
 
163
  for x2 in self._x2:
164
  try:
165
  if mu == 0.5:
166
+ mu_data = self.get_data_txt(mu, x1, x2)
167
  else:
168
+ mu_data = self.get_data(mu, x1, x2)
169
+ mu_data_t0 = mu_data[:-1,:,:]
170
  mu_data_t1 = mu_data[1:,:,:]
171
  data_t0.append(mu_data_t0)
172
  data_t1.append(mu_data_t1)
plotting.py ADDED
@@ -0,0 +1,174 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from fsi_reader import FsiDataReader
2
+ import matplotlib.pyplot as plt
3
+ import numpy as np
4
+ from matplotlib.tri import Triangulation
5
+ from matplotlib.animation import FuncAnimation
6
+ from scipy.interpolate import griddata
7
+
8
+ def single_plot(data, mesh_points):
9
+ data = np.squeeze(data) # Shape becomes (1317,)
10
+ print(data.shape)
11
+ print(mesh_points.shape)
12
+ x, y = mesh_points[:, 0], mesh_points[:, 1]
13
+
14
+ # Create figure with subplots
15
+ fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(16, 6),
16
+ gridspec_kw={'width_ratios': [1, 1.2]})
17
+
18
+ # Approach 1: Triangulation-based contour plot
19
+ tri = Triangulation(x, y)
20
+ contour = ax1.tricontourf(tri, data, levels=40, cmap='viridis')
21
+ fig.colorbar(contour, ax=ax1, label='Value', shrink=0.3)
22
+ ax1.set_title('Contour Plot of Field Data')
23
+ ax1.set_aspect('equal')
24
+
25
+ # Approach 2: Scatter plot with interpolated background
26
+ grid_x, grid_y = np.mgrid[x.min():x.max():100j, y.min():y.max():100j]
27
+ grid_z = griddata((x, y), data, (grid_x, grid_y), method='cubic')
28
+
29
+ im = ax2.imshow(grid_z.T, origin='lower', extent=[x.min(), x.max(),
30
+ y.min(), y.max()], cmap='plasma')
31
+ ax2.scatter(x, y, c=data, edgecolor='k', lw=0.3, cmap='plasma', s=15)
32
+ fig.colorbar(im, ax=ax2, label='Interpolated Value', shrink=0.3)
33
+ ax2.set_title('Interpolated Surface with Sample Points')
34
+
35
+ # Common formatting
36
+ for ax in (ax1, ax2):
37
+ ax.set_xlabel('X Coordinate')
38
+ ax.set_ylabel('Y Coordinate')
39
+ ax.grid(True, alpha=0.3)
40
+
41
+ plt.tight_layout()
42
+ plt.show()
43
+
44
+ def create_field_animation(data_frames, mesh_frames, interval=100, save_path=None):
45
+ """
46
+ Create an animation of time-varying 2D field data on a mesh.
47
+
48
+ Parameters:
49
+ -----------
50
+ data_frames : list of arrays
51
+ List of data arrays for each time frame (each with shape [1, 1317, 1] or similar)
52
+ mesh_frames : list of arrays or single array
53
+ Either a list of mesh coordinates for each frame or a single fixed mesh
54
+ interval : int
55
+ Delay between animation frames in milliseconds
56
+ save_path : str, optional
57
+ Path to save the GIF animation
58
+ """
59
+
60
+ plt.rcParams.update({
61
+ 'font.size': 20, # Base font size
62
+ 'axes.titlesize': 20, # Title font size
63
+ 'axes.labelsize': 20, # Axis label size
64
+ 'xtick.labelsize': 20, # X-tick label size
65
+ 'ytick.labelsize': 18, # Y-tick label size
66
+ 'figure.titlesize': 22 # Super title size (if used)
67
+ })
68
+
69
+ # Determine if mesh is fixed or time-varying
70
+ mesh_varying = isinstance(mesh_frames, list)
71
+
72
+ # Get initial mesh and data
73
+ mesh_initial = mesh_frames[0] if mesh_varying else mesh_frames
74
+ data_initial = np.squeeze(data_frames[0])
75
+
76
+ # Extract coordinates
77
+ x, y = mesh_initial[:, 0], mesh_initial[:, 1]
78
+
79
+ # Create figure
80
+ fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(50, 10),
81
+ gridspec_kw={'width_ratios': [1.5, 1.5]})
82
+
83
+ # Calculate global min/max for consistent colorbars
84
+ all_data = np.concatenate([np.squeeze(frame) for frame in data_frames])
85
+ vmin, vmax = all_data.min(), all_data.max()
86
+
87
+ # Create initial triangulation
88
+ tri_initial = Triangulation(x, y)
89
+
90
+ # Set up first subplot - contour
91
+ contour = ax1.tricontourf(tri_initial, data_initial, levels=40, cmap='viridis',
92
+ vmin=vmin, vmax=vmax)
93
+ # Add contour lines for better visibility
94
+ contour_lines = ax1.tricontour(tri_initial, data_initial, levels=15,
95
+ colors='black', linewidths=0.5, alpha=0.7)
96
+
97
+ fig.colorbar(contour, ax=ax1, label='Value', shrink=0.3)
98
+ ax1.set_title('Contour Plot of Field Data')
99
+ ax1.set_aspect('equal')
100
+
101
+ # Set up second subplot - interpolated surface with scatter points
102
+ grid_x, grid_y = np.mgrid[x.min():x.max():100j, y.min():y.max():100j]
103
+ grid_z = griddata((x, y), data_initial, (grid_x, grid_y), method='cubic')
104
+
105
+ im = ax2.imshow(grid_z.T, origin='lower', extent=[x.min(), x.max(),
106
+ y.min(), y.max()],
107
+ cmap='plasma', vmin=vmin, vmax=vmax)
108
+ scat = ax2.scatter(x, y, c=data_initial, edgecolor='k', lw=0.3,
109
+ cmap='plasma', s=15, vmin=vmin, vmax=vmax)
110
+
111
+ fig.colorbar(im, ax=ax2, label='Interpolated Value', shrink=0.3)
112
+ ax2.set_title('Interpolated Surface with Sample Points')
113
+
114
+ # Common formatting
115
+ for ax in (ax1, ax2):
116
+ ax.set_xlabel('X Coordinate')
117
+ ax.set_ylabel('Y Coordinate')
118
+ ax.grid(True, alpha=0.3)
119
+
120
+ # Add frame counter
121
+ time_text = ax1.text(0.02, 0.98, '', transform=ax1.transAxes,
122
+ fontsize=10, va='top', ha='left')
123
+
124
+ plt.tight_layout()
125
+
126
+ # Update function for animation
127
+ def update(frame):
128
+ # Get current data
129
+ data = np.squeeze(data_frames[frame])
130
+
131
+ # Get current mesh if varying
132
+ if mesh_varying:
133
+ mesh = mesh_frames[frame]
134
+ x, y = mesh[:, 0], mesh[:, 1]
135
+ tri = Triangulation(x, y)
136
+ else:
137
+ mesh = mesh_frames
138
+ x, y = mesh[:, 0], mesh[:, 1]
139
+ tri = tri_initial
140
+
141
+ # Update contour plot
142
+ for c in ax1.collections:
143
+ c.remove()
144
+ new_contour = ax1.tricontourf(tri, data, levels=40, cmap='viridis',
145
+ vmin=vmin, vmax=vmax)
146
+ new_lines = ax1.tricontour(tri, data, levels=15, colors='black',
147
+ linewidths=0.5, alpha=0.7)
148
+
149
+ # Update interpolated surface
150
+ grid_z = griddata((x, y), data, (grid_x, grid_y), method='cubic')
151
+ im.set_array(grid_z.T)
152
+
153
+ # Update scatter points
154
+ scat.set_offsets(mesh)
155
+ scat.set_array(data)
156
+
157
+ # Update frame counter
158
+ time_text.set_text(f'Frame: {frame+1}/{len(data_frames)}')
159
+
160
+ return [new_contour, new_lines, im, scat, time_text]
161
+
162
+ # Create animation
163
+ anim = FuncAnimation(fig, update, frames=len(data_frames),
164
+ interval=interval, blit=False)
165
+
166
+ # Save if path provided
167
+ if save_path:
168
+ print(f"Saving animation to {save_path}...")
169
+ if save_path.endswith('.gif'):
170
+ anim.save(save_path, writer='pillow', dpi=150)
171
+ else:
172
+ anim.save(save_path, writer='ffmpeg', dpi=150)
173
+
174
+ return anim