Exercise 1¶

In [13]:

# Read pile data from csv-file
df_piles = pd.read_csv('piles.csv')

# Read steel profile data from csv-file
df_profiles = pd.read_csv('steel_profiles.csv')

# Merge dataframes on "Profile" column (similar to Excel VLOOKUP)
df_merged = df_piles.merge(df_profiles, on='Profile', how='left')

In [15]:

# Display first five rows of dataframe of steel profiles
df_profiles.head()

Out[15]:

	Profile	h[mm]	b[mm]	Iy[mm4]	Wel_y[mm3]	g[kg/m]
0	HE100A	96	100	3490000	72.8	16.7
1	HE120A	114	120	6060000	106.0	19.9
2	HE140A	133	140	10300000	155.0	24.7
3	HE160A	152	160	16700000	220.0	30.4
4	HE180A	171	180	25100000	294.0	35.5

In [14]:

# Display dataframe of piles
df_piles

Out[14]:

	Pile_type	Profile
0	P01	HE200A
1	P20	HE220A
2	P05	HE240B
3	P23	NaN
4	P04	HE200A
5	P01	HE300B

In [16]:

# Display merged dataframe 
df_merged

Out[16]:

	Pile_type	Profile	h[mm]	b[mm]	Iy[mm4]	Wel_y[mm3]	g[kg/m]
0	P01	HE200A	190.0	200.0	36900000.0	389.0	42.3
1	P20	HE220A	210.0	220.0	54100000.0	515.0	50.5
2	P05	HE240B	240.0	240.0	112600000.0	938.0	83.2
3	P23	NaN	NaN	NaN	NaN	NaN	NaN
4	P04	HE200A	190.0	200.0	36900000.0	389.0	42.3
5	P01	HE300B	300.0	300.0	251700000.0	1680.0	117.0

Exercise 2¶

In [55]:

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Set filename for dataset of forces from an IBDAS shell element
filename = 'Crack_width_Seg7_y_direction.csv'

# Read CSV-file
df = pd.read_csv(filename, skip_blank_lines=True)

# Filter dataframe for load case and criterion for critical combination
criterion = 'max My'
df = df[df['Criterion'] == criterion]

df.head()

Out[55]:

	Criterion	x[m]	y[m]	N[kN/m]	M[kNm/m]	w_k[mm]
2	max My	22.115	7.603	-459.6	66.1	0.002553
5	max My	22.250	7.603	-432.1	87.8	0.013417
8	max My	22.250	7.747	-443.1	80.9	0.009092
11	max My	22.115	7.747	-497.0	69.0	0.002068
14	max My	22.115	8.159	-522.5	56.7	0.000000

In [56]:

# Round dataframe coordinate columns
df['x[m]'] = round(df['x[m]'], 1)
df['y[m]'] = round(df['y[m]'], 1)

# Pivot data frame data into matrix form for heatmap plotting
pivot_final = df.pivot(index='y[m]', columns='x[m]', values='w_k[mm]').sort_index(ascending=False)

# Set max allowable crack with for concrete slab for use as max value of colorbar
vmax = 0.3

# Create figure
plt.figure(figsize=(18, 8))

# Plot heatmap wiht annotation
sns.heatmap(pivot_final, annot=True, annot_kws={'size': 10}, vmax=vmax,
                  fmt=".2f", square=True, cbar_kws={"orientation": "horizontal"}, cmap='Reds')

# Set titles and axes labels
plt.title(f'Crack width for Segment 7, Criterion: {criterion}', fontsize=20)
plt.xlabel('Global x-coordinate [m]', fontsize=16)
plt.ylabel('Global y-coordinate [m]', fontsize=16)

Out[56]:

Text(142,0.5,'Global y-coordinate [m]')

Some extra stuff - Conditional coloring of values¶

In [61]:

# Create figure
plt.figure(figsize=(18, 8))

# Plot heatmap wiht annotation, save axes object so it can be accessed later
ax = sns.heatmap(pivot_final, annot=True, annot_kws={'size': 10}, vmax=vmax,
                  fmt=".2f", square=True, cbar_kws={"orientation": "horizontal"}, cmap='Reds')

# --- Color values that exceed the max value ---

# Loop over all annotations of the axes object
for annot in ax.texts:
    
    # Annot will now be an object which prints 'Text(x, y, w_k)'
    
    # Extract the crack width part of the Text object and convert from string to float 
    wk = float(annot.get_text())
    
    # Set all values that exceed vmax to bold and a special color
    if wk > vmax: 
        annot.set_weight('bold')
        annot.set_color('cyan')
        annot.set_size(12)

plt.show()

In [ ]:

485 KiB Raw Blame History

Exercise 1¶

Exercise 2¶

Some extra stuff - Conditional coloring of values¶

485 KiB

Raw Blame History