Skip to main content

A package for dynamical systems analysis

Project description

dynlab

Dynlab is a python package to make dynamical systems analysis faster and easier, by providing practitioners prebuilt modules for calculating common diagnostics. Dynlab currently offers several easy to use diagnostics for analyzing your dynamical systems including an FTLE calculator FLTE, an attraction and repulsion rate calculator AttractionRate, and a trajectory repulsion rate calculator TrajectoryRepulsionRate. Dynlab also offers users a variety of prewritten flows to experiment and play around with, including the double gyre double_gyre, the Duffing oscillator duffing_oscillator, the pendulum with damping and forcing pendulum, the Van Der Pol oscillator van_der_pol_oscillator, the Lotka-Voltera flow lotka_volterra, the Lorenz system lorenz, and many more. Note that all the flows are written in such a way that they can easily be passed to an ode integrator and as such even the autonomous flows will still require a time parameter to be passed to them.

Currently supported Lagrangian diagnostics are:
Finite-Time Lyapunov Exponent: FTLE
Lagrangian Coherent Structures: LCS

Currently supported Eulerian diagnostics are:
Attraction Rate: AttractionRate
Repulsion Rate: RepulsionRate
infinitesimal Lyapunov ExponentS: iLES
Trajectory Repulsion Rate: TrajectoryRepulsionRate
Trajectory Repulsion Ratio: TrajectoryRepulsionRatio

This package is readily available on pypi and can be easily installed with the command pip install dynlab. Note that dynlab is written with python 3.11 and you may need to update your python installation to take advantage of it.

Example

import matplotlib.pyplot as plt
from dynlab.diagnostics import FTLE
from dynlab.flows import double_gyre
x = np.linspace(0, 2, 101)
y = np.linspace(0, 1, 101)
ftle = FTLE(num_threads=1).compute(x, y, double_gyre, (10, 0), edge_order=2, rtol=1e-8, atol=1e-8)
plt.pcolormesh(x, y, ftle, shading='gouraud')

alt text

import matplotlib.pyplot as plt
from dynlab.diagnostics import LCS
from dynlab.flows import double_gyre
x = np.linspace(0, 2, 201)
y = np.linspace(0, 1, 101)
lcs = LCS()
attracting_lcs = lcs.compute(x, y, f=double_gyre, t=(10, 0), percentile=80)
for line in attracting_lcs:
    plt.plot(line[:, 0],line[:, 1], 'b')

alt text

import matplotlib.pyplot as plt
from dynlab.diagnostics import iLES
from dynlab.flows import double_gyre
x = np.linspace(0, 2, 201)
y = np.linspace(0, 1, 101)
attracting_iles = iLES().compute(x, y, f=double_gyre, t=0, kind='attracting', force_eigenvectors=True)
repelling_iles = iLES().compute(x, y, f=double_gyre, t=0, kind='repelling', force_eigenvectors=True)

xx, yy = np.meshgrid(x[::10],y[::10])
u, v = double_gyre(0, (xx, yy))
plt.quiver(xx,yy,u,v)
for line in attracting_iles:
    plt.plot(line[:, 0],line[:, 1], 'b', linewidth=3)
for line in repelling_iles:
    plt.plot(line[:, 0],line[:, 1], 'r', linewidth=3)

alt text

import matplotlib.pyplot as plt
from dynlab.diagnostics import AttractionRate
from dynlab.flows import double_gyre
x = np.linspace(0, 20000, 101)
y = np.linspace(-4000, 4000, 101)
u, v = bickley_jet(0, np.meshgrid(x, y))
attraction_rate = AttractionRate().compute(x, y, u=u, v=v, edge_order=2)
# note that lower values of the attraction rate field equate to higher levels of attraction
# so we'll plot the negative of the attraction rate field to highlight areas of greatest attraction.
plt.pcolormesh(x, y, -attraction_rate, shading='gouraud')

alt text

import matplotlib.pyplot as plt
from dynlab.diagnostics import Rhodot
from dynlab.flows import bead_on_a_rotating_hoop
x = np.linspace(-1, 1, 401)*3
y = np.linspace(-1, 1, 401)*2.5
rhodot, nudot = TrajectoryRepulsionRate().compute(x, y, f=bead_on_a_rotating_hoop, t=0, edge_order=2)
plt.pcolormesh(x, y, rhodot, shading='gouraud')

alt text

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

dynlab-0.1.0.tar.gz (25.5 kB view hashes)

Uploaded Source

Built Distribution

dynlab-0.1.0-py3-none-any.whl (26.9 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page