Time machine - Extended period simulations with OOPNET¶
In this example we will discuss performing extended period simulations with OOPNET.
As always, we have to import all required packages. This time, we will also need datetime.timedelta to specify some time related settings in the model.
import os
from datetime import timedelta
from matplotlib import pyplot as plt
import oopnet as on
Next, we read the “C-Town” model:
filename = os.path.join('data', 'C-town.inp')
net = on.Network.read(filename)
Now, we change the simulation duration to 6 hours and the reporting time step to 5 minutes. For this, we use datetime.timedelta:
net.times.duration = timedelta(hours=6)
net.times.reporttimestep = timedelta(minutes=5)
Wen can now simulate the model and interact with the oopnet.SimulationReport. When we access a specific property of this report, we now get a pandas.DataFrame object instead of a pandas.Series like we get when doing steady-state simulations:
rpt = net.run()
print(rpt.pressure)
id J511 J411 J414 ... T5 T2 T4
time ...
2016-01-01 00:00:00 28.83 64.42 39.00 ... 2.25 2.95 2.35
2016-01-01 00:05:00 28.71 64.40 38.97 ... 2.20 2.95 2.23
2016-01-01 00:10:00 28.59 64.37 38.95 ... 2.15 2.95 2.12
2016-01-01 00:15:00 28.48 64.35 38.93 ... 2.10 2.95 2.00
2016-01-01 00:20:00 28.36 64.32 38.90 ... 2.05 2.95 1.88
... ... ... ... ... ... ... ...
2016-01-01 05:40:00 -7520000.00 63.65 38.24 ... -0.00 2.31 -0.00
2016-01-01 05:45:00 -7530000.00 63.63 38.22 ... -0.00 2.30 -0.00
2016-01-01 05:50:00 -7530000.00 63.62 38.20 ... -0.00 2.29 -0.00
2016-01-01 05:55:00 -7530000.00 63.60 38.18 ... -0.00 2.28 -0.00
2016-01-01 06:00:00 -9960000.00 63.29 37.88 ... -0.00 2.27 -0.00
[73 rows x 396 columns]
Using pandas, we can also get some basic statistics:
print(rpt.pressure.describe())
id J511 J411 J414 ... T5 T2 T4
count 7.300000e+01 73.000000 73.000000 ... 73.000000 73.000000 73.000000
mean -5.704922e+06 64.220137 38.803562 ... 0.863288 2.698767 0.359315
std 4.155719e+06 0.268388 0.266418 ... 0.696533 0.237843 0.659588
min -1.090000e+07 63.290000 37.880000 ... -0.000000 2.270000 -0.000000
25% -8.130000e+06 64.170000 38.760000 ... 0.240000 2.490000 0.000000
50% -7.530000e+06 64.300000 38.880000 ... 0.730000 2.720000 0.000000
75% 2.732000e+01 64.390000 38.970000 ... 1.420000 2.950000 0.430000
max 2.883000e+01 64.560000 39.150000 ... 2.250000 2.950000 2.350000
[8 rows x 396 columns]
Now, we want to create a plot using this data. In the plot, we want to show the flow through pipe P1000 on one subplot, while showing the pressure at junctions J10 and J1058 on a second subplot. For this, we first have to create a new matplotlib matplotlib.Figure.
fig = plt.figure(1)
We then create an matplotlib.Axes object for the first plot, plot the data, set labels for the axes and add a legend.
rpt = net.run()
print(rpt.pressure)
print(rpt.pressure.describe())
Next, we do the same with the pressure data. Note, that we don’t have to create a legend because pandas does it automatically when plotting several lines.
ax2 = fig.add_subplot(212)
rpt.pressure[['J10', 'J1058']].plot(ax=ax2)
plt.xlabel('time', fontsize=16)
plt.ylabel('p (m)', fontsize=16)
Finally, we can show our plots:
plt.show()
Summary¶
import os
from datetime import timedelta
from matplotlib import pyplot as plt
import oopnet as on
filename = os.path.join('data', 'C-town.inp')
net = on.Network.read(filename)
net.times.duration = timedelta(hours=6)
net.times.reporttimestep = timedelta(minutes=5)
rpt = net.run()
print(rpt.pressure)
print(rpt.pressure.describe())
fig = plt.figure(1)
ax1 = fig.add_subplot(211)
rpt.flow['P1000'].plot(ax=ax1)
plt.xlabel('time', fontsize=16)
plt.ylabel('Q (l/s)', fontsize=16)
plt.legend()
ax2 = fig.add_subplot(212)
rpt.pressure[['J10', 'J1058']].plot(ax=ax2)
plt.xlabel('time', fontsize=16)
plt.ylabel('p (m)', fontsize=16)
plt.show()