def test_efficiency_calibration(self):
"""Do efficiency calibration of detector to spectrum of calibrated 152-Eu source"""
# test are run sorted by the string representations of the test methods; we need this to be run first
if self.energy_calibration is None:
self.test_energy_calibration()
# generate expected peaks from source specs
Eu152_expected = source_to_dict(self.Eu152_source_specs, info='lines')
efficiency_calib_bkg = sp.interpolate_bkg(
counts=self.Eu152_spectrum[1])
efficiency_calib_peaks = sp.fit_spectrum(
counts=self.Eu152_spectrum[1],
energy_cal=self.energy_calibration['func'],
expected_accuracy=self.
accuracy, # self.energy_calibration['accuracy']
expected_peaks=Eu152_expected,
bkg=efficiency_calib_bkg)
# check whether all peaks have been found
self.assertListEqual(sorted(efficiency_calib_peaks.keys()),
sorted(Eu152_expected.keys()))
self.efficiency_calibration = sp.do_efficiency_calibration(
observed_peaks=efficiency_calib_peaks,
source_specs=self.Eu152_source_specs)
# 50 % error on fit parameters
self.assertTrue(
all(self.efficiency_calibration['perr'] /
self.efficiency_calibration['popt'] <= 0.5))
def test_dose(self):
"""Testing the dose calculations on the example spectrum"""
# test are run sorted by the string representations of the test methods; we need this to be run first
if self.energy_calibration is None:
self.test_energy_calibration()
# test are run sorted by the string representations of the test methods; we need this to be run first
if self.efficiency_calibration is None:
self.test_efficiency_calibration()
peaks_sample, bkg_sample = sp.fit_spectrum(
counts=self.sample_spectrum[1],
energy_cal=self.energy_calibration['func'],
efficiency_cal=self.efficiency_calibration['func'],
t_spec=self.t_sample)
dose = sp.get_dose(peaks_sample, distance=50, time=2000)
self.assertTrue(np.isclose(dose['nominal'], 1.1330667863561383))
self.assertTrue(np.isclose(dose['sigma'], 0.03766586624214202))
self.assertTrue(dose['unit'] == 'uSv')
dose = sp.get_dose(peaks_sample, distance=1)
self.assertTrue(np.isclose(dose['nominal'], 1.8437455480798448))
self.assertTrue(np.isclose(dose['sigma'], 0.05874337083851746))
self.assertTrue(dose['unit'] == 'uSv/h')
selected_peaks = select_peaks(['65_Zn', '48_V', '7_Be'], peaks_sample)
self.assertEqual(len(selected_peaks), 4)
selected_dose = sp.get_dose(selected_peaks, distance=50, time=2000)
self.assertTrue(
np.isclose(selected_dose['nominal'], 0.5343614796430098))
selected_dose_time_1 = sp.get_dose(selected_peaks,
distance=50,
time=1e6)['nominal']
selected_dose_time_2 = sp.get_dose(selected_peaks,
distance=50,
time=1e7)['nominal']
self.assertTrue(np.isclose(selected_dose_time_1, selected_dose_time_2))
def test_energy_calibration(self):
"""Do energy calibration of detector channels to spectrum of 152-Eu source"""
energy_calib_bkg = sp.interpolate_bkg(counts=self.Eu152_spectrum[1])
energy_calib_peaks = sp.fit_spectrum(
counts=self.Eu152_spectrum[1],
expected_peaks=self.energy_calib_peaks['channel'],
bkg=energy_calib_bkg)
# check whether all peaks have been found
self.assertListEqual(sorted(energy_calib_peaks.keys()),
sorted(self.energy_calib_peaks['channel'].keys()))
self.energy_calibration = sp.do_energy_calibration(
observed_peaks=energy_calib_peaks,
peak_energies=self.energy_calib_peaks['energy'])
# 50 % error on fit parameters
self.assertTrue(
all(self.energy_calibration['perr'] /
self.energy_calibration['popt'] <= 0.5))
def test_benchmark(self):
"""
Do a benchmark of the software with two, calibrated sources; 22-Na and 133-Ba. The expected peaks of both spectra
are fitted to first benchmark the energy calibration with the peak positions. Then the integrated and scaled activity
is compared to the theoretically expected activity to benchmark the efficiency calibration as well as the fitting and
background subtraction.
Overall, a reconstruction accuracy of above 90 % is achieved for both sources (after approx. 10 minute measurement)
The 22-Na source has one line/peak => 22-Na activity is (93.4 +- 2.7)% accurately measured
The 133-Ba source has 5 lines/peaks => 133-Ba activity is (97.0 +- 1.2)% accurately measured
"""
# test are run sorted by the string representations of the test methods; we need this to be run first
if self.energy_calibration is None:
self.test_energy_calibration()
# test are run sorted by the string representations of the test methods; we need this to be run first
if self.efficiency_calibration is None:
self.test_efficiency_calibration()
# test energy calibration by applying to Na22 and Ba133 spectra
# generate expected peaks from source specs
Na22_expected = source_to_dict(self.Na22_source_specs, info='lines')
# fit spectrum of source
Na22_peaks, Na22_bkg = sp.fit_spectrum(
counts=self.Na22_spectrum[1],
energy_cal=self.energy_calibration['func'],
efficiency_cal=self.efficiency_calibration['func'],
t_spec=self.t_Na22,
expected_accuracy=self.accuracy
) # self.energy_calibration['accuracy']
# check whether all expected peaks have been found from library
self.assertTrue(all(ep in Na22_peaks for ep in Na22_expected.keys()))
# check whether all activities are in Bq
self.assertTrue(
all('normalized' == Na22_peaks[p]['activity']['type']
for p in Na22_peaks))
# check whether all activities are calibrated for efficiency
self.assertTrue(
all(Na22_peaks[p]['activity']['calibrated'] for p in Na22_peaks))
# check whether all peaks are at correct energies within 1 per mille accuracy
for na22_peak in Na22_peaks:
if na22_peak in Na22_expected:
low, high = (x * Na22_expected[na22_peak]
for x in (1 - self.accuracy, 1 + self.accuracy))
self.assertTrue(
low <= Na22_peaks[na22_peak]['peak_fit']['popt'][0] <= high
)
# check for correct activity from library
Na22_activity_meas = sp.get_activity(Na22_peaks)
Na22_activity_theo = decay_law(
t=self.Na22_source_specs['timestamp_measurement'] -
self.Na22_source_specs['timestamp_calibration'],
x0=np.array(self.Na22_source_specs['activity']),
half_life=self.Na22_source_specs['half_life'])
# check to see at least 90% of the expected activity; only order of magnitude relevant
self.assertTrue(0.9 <= Na22_activity_meas['22_Na']['nominal'] /
Na22_activity_theo[0] <= 1.0)
# generate expected peaks and probabilities from source specs
Ba133_expected = source_to_dict(self.Ba133_source_specs, info='lines')
# fit spectrum of source
Ba133_peaks, Ba133_bkg = sp.fit_spectrum(
counts=self.Ba133_spectrum[1],
energy_cal=self.energy_calibration['func'],
efficiency_cal=self.efficiency_calibration['func'],
t_spec=self.t_Ba133,
expected_accuracy=self.accuracy
) # self.energy_calibration['accuracy']
# check whether all expected peaks have been found from library
self.assertTrue(all(ep in Ba133_peaks for ep in Ba133_expected.keys()))
# check whether all activities are in Bq
self.assertTrue(
all('normalized' == Ba133_peaks[p]['activity']['type']
for p in Ba133_peaks))
# check whether all activities are calibrated for efficiency
self.assertTrue(
all(Ba133_peaks[p]['activity']['calibrated'] for p in Ba133_peaks))
# check whether all peaks are at correct energies within 1 per mill accuracy
for ba133_peak in Ba133_peaks:
if ba133_peak in Ba133_expected:
low, high = (x * Ba133_expected[ba133_peak]
for x in (1 - self.accuracy, 1 + self.accuracy))
self.assertTrue(
low <= Ba133_peaks[ba133_peak]['peak_fit']['popt'][0] <=
high,
msg=str(self.energy_calibration['accuracy']))
# check for correct activity
Ba133_activity_meas = sp.get_activity(Ba133_peaks)
Ba133_activity_theo = decay_law(
t=self.Ba133_source_specs['timestamp_measurement'] -
self.Ba133_source_specs['timestamp_calibration'],
x0=np.array(self.Ba133_source_specs['activity']),
half_life=self.Ba133_source_specs['half_life'])
# check to see at least 90% of the expected activity; only order of magnitude relevant
self.assertTrue(0.9 <= Ba133_activity_meas['133_Ba']['nominal'] /
Ba133_activity_theo[0] <= 1.0)
