Result Objects#

class SimulationResults#

Container object that encloses all the results from a single simulation. This object enables you to retrieve output states for Bodies, Requests and Rvs for each output time step.

There are two ways to instruct msolve to store the simulation results. In the Model.simulate method, one of the following must be set:

onOutputStep
Function that is called when the solver hits an output step. It can be used for animation and live plots.
returnResults
Boolean that if True, causes SimulationResults to be populated.

asDataFrame(start=0, end=None)#

Returns the runData object as a dictionary of Pandas dataframes. The keys of the dictionary are the objects whose results are stored.

Parameters:

start (float) – Start time value for trimming each DataFrame.
end (float) – End time value for trimming each DataFrame.

Returns:

A new dictionary where each value is the result of the ResultObject.getDataFrame method called on the original value.

getObject(obj)#

Extracts the specified results from the SimulationResults (run) container and passes them into a ResultObject object.

Parameters:: obj (Body, Request, Rv, Name) – The object whose results are to be retrieved, or the unique name given to the object.
Returns:: The ResultObject that is associated to the obj.

Note

Using the object as a lookup key is useful, so you don’t have to know the name of the object. This approach allows for convenient lookups without relying on object names.

Using the name string as a lookup key is advantageous in scenarios such as plot templates or when working with serialized data in formats like pickle files. In such cases, the model instance and its associated objects may not be directly accessible, making the use of objects impractical. Using the name string allows for reliable lookups even when the object hierarchy is not available.

class SimulationResultsHistory(**kwds)#

A list that can be used to store, serialize and de-serialize SimulationResults for cross plotting. Any simulation executed with the flag store=True will append an instance of the current SimulationResults to this class. The SimulationResultsHistory is stored as part of the model.

Example

from msolve import *
model = createDemoPendulum()
# run two simulations, each will be stored.
model.simulate(type="TRANSIENT", end=1, steps=100, store=True)
model.simulate(type="TRANSIENT", end=5, steps=1000, store=True)

# print each node in the list as a Pandas DataFrame
for run in model.simulationResultsHistory:
  print(run.asDataFrame())

serialize_json(filename=None)#: Serializes the SimulationResultsHistory to JSON. If a filename is provided, the data is written to a JSON file. If not, the method returns a JSON string.

serialize_pickle(filename=None, serialize_all=False)#: Pickles the SimulationResultsHistory object. If filename is defined, the data is written to a pkl file. If not, the method returns a pickle string. If serialize_all=False, only the tagged (named) result objects will be serialized. This is done for memory saving purposes.

class ResultObject#

Base class for all the result objects of a simulation. The methods of this class are the access point to the numerical data of the simulation results.

getComponent(component)#

Queries a ResultObject and returns the time history of a specific component of the object. The components can be indexed with their index number or their respective label.

The available components for each object are listed in BodyResult, FlexBodyResult, PointMassResult, RequestResult, RvResult, FrfBodyResult, FrfRequestResult.

getDataFrame()#

The entire ResultObject corresponding to the owner is returned as a Pandas DataFrame. It contains all the components and uses labels as column names.

Useful Pandas DataFrame functionalities:

df.info(): Prints concise summary of the DataFrame.
df.describe(): Generates descriptive statistics for all columns including min, std, max, mean.
df[‘column_name’] or df.column_name: Returns the column with name ‘column_name’ as a pandas Series.
df.columns: Prints a pandas index corresponding to all the columns.

getStep(step=-1)#: Queries a ResultObject and returns its data for the specific time step or the closest previous step. step can be an integer or a float value corresponding to the time step.

class BodyResult#

Contains all the results associated with a single Part of the model.

At each output time, the following results are stored:

Components	Label	Index
Local Part Reference Frame (LPRF) position	X, Y, Z	0 - 2
Euler parameters	E0, E1, E2, E3	3 - 6

Example

Extract the simulation results of a Body with getComponent.#

from msolve import *

model = Model(output='body_results')

Units(system='MKS')
Accgrav(kgrav=-9.81)

ground = Part(ground=True)
global_ref = Marker(part=ground)

ball = Part(mass=1, ip=[1]*3, cm=Marker(qp=[0,0,10], zv=[0,0,1]))

run = model.simulate(type='TRANSIENT', end=2, dtout=0.01, returnResults=True)

body_results = run.getObject(ball)
#assert type(body_results) == BodyResult

time = body_results.times
ball_disp  = body_results.getComponent('Z')

import matplotlib.pyplot as plt
plt.plot(time, ball_disp)
plt.title('Vertical Displacement')
plt.xlabel('time (s)')
plt.ylabel('z (m)')
plt.grid()
plt.show()

class FlexBodyResult#

Contains results associated with a single FlexBody of the model.

At each output time, the following results are stored:

Components	Label	Index
Center of mass position	X, Y, Z	0 - 2
Euler parameters	E0, E1, E2, E3	3 - 6
Strain Energy	SE	7

Example

For this example, a mtx file is required. The file is located in the mbd_modeling\flexbodies folder in the MotionSolve tutorials Model Files. You may copy the file to your working directory.

Extract the simulation results of a FlexBody with asDataFrame().#

from msolve import *

model = Model(output='flex_results')

ground = Part(ground=True)
global_ref = Marker(part=ground)
Units(system='MKS')
Accgrav(kgrav=-9.81)

flex = FlexBody(mtx_file="sla_flex_left.mtx", qg=[0,0,4])
flex_marker = Marker(flex_body=flex, qp=flex.qg, zv=[0,0,1])

vel_request = Request(type="VELOCITY", i=flex_marker, j=global_ref)

run = model.simulate (type="DYNAMIC", end=0.5, steps=500, returnResults=True)

results_dict = run.asDataFrame()
print(results_dict.keys())

flex_df        = results_dict[flex]
vel_request_df = results_dict[vel_request]
print(vel_request_df.columns)

import matplotlib.pyplot as plt
plt.plot(flex_df['Z'], label="flex cm z-displacement")
plt.plot(abs(vel_request_df['VZ']), label="flex lprf z-velocity")
plt.xlabel('time (s)')
plt.legend()
plt.grid()
plt.show()

class FrfBodyResult#: Contains all the results associated with a single Part of the model in a FrequencyResponse simulation for Rigid, Flexible and PointMass bodies.

class PointMassResult#

Contains all the results associated with a single PointMass of the model.

At each output time, the following results are stored:

Components	Label	Index
LPRF position	X, Y, Z	0 - 2

Example

Extract the simulation results of a PointMass with getStep().#

from msolve import *

model = Model(output='pmass_results')

Units(system='MKS')
Accgrav(kgrav=-9.81)

ground = Part(ground=True)
global_ref = Marker(part=ground)

pmass = PointMass(mass=10, cm=Marker(qp=[0,0,10], zv=[0,0,1]), vz=8.0)

run = model.simulate(type='DYNAMIC', end=2, dtout=0.01, returnResults=True)

pmass_results = run.getObject(pmass)
#assert type(pmass_results) == PointMassResult

time = pmass_results.times

import matplotlib.pyplot as plt

# loop over the number of output time steps
for i in range(1,200):
  pmass_disp  = pmass_results.getStep(i)[2]

  plt.scatter(i*0.01, pmass_disp, linewidths=1.0)

plt.title('Vertical Displacement')
plt.xlabel('time (s)')
plt.ylabel('z (m)')
plt.grid()
plt.show()

class RequestResult#

Contains all the results associated with a single Request of the model.

At each output time, the following results are stored:

Components	Label	Index
Displacement Request	MAG, X, Y, Z, null, PSI, THETA, PHI	0 - 7
Velocity Request	VM, VX, VY, VZ, WM, WX, WY, WZ	0 - 7
Acceleration Request	ACCM, ACCX, ACCY, ACCZ, WDTM, WDTX, WDTY, WDTZ	0 - 7
Force Request	FM, FX, FY, FZ, TM, TX, TY, TZ	0 - 7
Expression Request	F1, F2, F3, F4, F5, F6, F7, F8	0 - 7

Example

Extract the simulation results of a Request with getDataFrame().#

from msolve import *

model = Model(output='request_results')

Units(system='mmks')
Accgrav(kgrav=-9.81)

ground = Part(ground=True)
global_ref = Marker(part=ground)

part = Part(mass=10, ip=[1e3]*3, cm=Marker(qp=[0,0,10], zv=[0,0,1]))

spdp = SpringDamper(type   = 'TRANSLATION',
                    i      = part.cm,
                    j      = global_ref,
                    k      = 0.05,
                    c      = 0.005,
                    force  = 50,
                    length = 10,
                    )

vel_request = Request(type='VELOCITY', i=part.cm, j=global_ref)
acc_request = Request(f1=f'ACCZ({part.cm.id},{global_ref.id})')

run1 = model.simulate(type='DYNAMIC', end=5, dtout=0.01, returnResults=True)

spdp.c = 0.05

run2 = model.simulate(type='DYNAMIC', end=8, dtout=0.01, returnResults=True)

vel_results = run1.getObject(vel_request)
acc_results = run1.getObject(acc_request)

df = vel_results.getDataFrame()
df.columns

time          = acc_results.times
acceleration  = acc_results.getComponent('F1')

import matplotlib.pyplot as plt

plt.plot(df.VZ)
plt.plot(time, acceleration)
plt.title('Part Velocity & Acceleration time-histories')
plt.legend(['Velocity', 'Acceleration'])
plt.xlabel('time (s)')
plt.grid()
plt.show()

class RvResult#

Contains all the results associated with a single Rv of the model.

At each output time, the following results are stored:

Components	Label	Index
Value of the integral of the Rv	RVAL	0
Value of the Rv	RVAL1	1

Example

Extract the simulation results of a Response variable (Rv).#

from msolve import *

model = Model(output='rv_results')

ground = Part(ground=True)
global_ref = Marker(part=ground)
Units(system='mks')
Accgrav(kgrav=-10)

part = Part(mass=100, ip=[1,1,1], cm = Marker(qp=[0,0,1]))

c = Dv (b=0.1, blimit=[0.01, 1])
spdp = SpringDamper (type="TRANSLATION", i=part.cm, j=global_ref, k=1, c=c)

rv = Rv(function=f"DZ({part.cm.id})")

run = model.simulate(type="STATIC", end=0.5, steps=20, returnResults=True, dsa='AUTO')

rv_output   = run.getObject(rv)
time     = rv_output.times
rv_rval  = rv_output.getComponent('RVAL')
rv_rval1 = rv_output.getComponent('RVAL1')

import matplotlib.pyplot as plt
plt.plot(time, rv_rval,  label="rval")
plt.plot(time, rv_rval1, label="rval1")
plt.xlabel('time (s)')
plt.legend()
plt.grid()
plt.show()