Showcase

If required, install pyuff and matplotlib.

#!!pip install pyuff
#!!pip install matplotlib

import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline

To analyse UFF file we first load the uff module and example file:

import pyuff
uff_file = pyuff.UFF('data/beam.uff')

First we can check which datasets are written in the file:

uff_file.get_set_types()

we see that first 4 datasets are 151: Header, 164: Units, 2420: Coordinate Systems and 2411: Nodes - Double Precision. Next we have several datasets 58 containing measurement data. To check what is written in the header (first dataset) use:

uff_file.read_sets(0)

We see that each dataset consists number of dictionary-like keys. We read and write directly to keys.

Reading from the UFF file

To load all datasets from the UFF file to data object use:

data = uff_file.read_sets()

The first dataset 58 (this is the fifth in the example file) contains following keys:

data[4].keys()

Most important keys are x: x-axis and data: y-axis that define the stored response:

plt.semilogy(data[4]['x'], np.abs(data[4]['data']))
plt.xlabel('Frequency  [Hz]')
plt.ylabel('FRF Magnitude [dB m/N]')
plt.xlim([0,1000])
plt.show()

Writing measurement data to UFF file

Here you can see a minimal working example for writing measured accelerance FRF data to the UFF file. First we load the accelerances:

measurement_point_1 = np.genfromtxt('data/meas_point_1.txt', dtype=complex)
measurement_point_2 = np.genfromtxt('data/meas_point_2.txt', dtype=complex)
measurement_point_3 = np.genfromtxt('data/meas_point_3.txt', dtype=complex)

measurement_point_1[0] = np.nan*(1+1.j)

measurement = [measurement_point_1, measurement_point_2, measurement_point_3]

In the next step we create a UFF file where we add dataset 58 for measurement consisting of the dictionary-like keys containing the measurement data and the information about the mesurement.

for i in range(3):
    print('Adding point {:}'.format(i + 1))
    response_node = 1
    response_direction = 1
    reference_node = i + 1
    reference_direction = 1
    acceleration_complex = measurement[i]
    frequency = np.arange(0, 1001)
    name = 'TestCase'
    data = {'type':58,
            'func_type': 4,
            'rsp_node': response_node,
            'rsp_dir': response_direction,
            'ref_dir': reference_direction,
            'ref_node': reference_node,
            'data': acceleration_complex,
            'x': frequency,
            'id1': 'id1',
            'rsp_ent_name': name,
            'ref_ent_name': name,
            'abscissa_spacing':1,
            'abscissa_spec_data_type':18,
            'ordinate_spec_data_type':12,
            'orddenom_spec_data_type':13}
    uffwrite = pyuff.UFF('./data/measurement.uff')
    uffwrite._write_set(data,'add')

Or we can use support function prepare_58 to prepare the dictionary for creating the UFF file. Functions for other datasets can be found in Supported datasets

for i in range(3):
    print('Adding point {:}'.format(i + 1))
    response_node = 1
    response_direction = 1
    reference_node = i + 1
    reference_direction = 1
    acceleration_complex = measurement[i]
    frequency = np.arange(0, 1001)
    name = 'TestCase'
    pyuff.prepare_58(func_type=4,
                rsp_node=response_node,
                rsp_dir=response_direction,
                ref_dir=reference_direction,
                ref_node=reference_node,
                data=acceleration_complex,
                x=frequency,
                id1='id1',
                rsp_ent_name=name,
                ref_ent_name=name,
                abscissa_spacing=1,
                abscissa_spec_data_type=18,
                ordinate_spec_data_type=12,
                orddenom_spec_data_type=13)