3.12

LeBailFit v1

../_images/LeBailFit-v1_dlg.png

LeBailFit dialog.

Summary

Do LeBail Fit to a spectrum of powder diffraction data.

Properties

Name Direction Type Default Description
InputWorkspace Input MatrixWorkspace Mandatory Input workspace containing the data to fit by LeBail algorithm.
OutputWorkspace Output Workspace2D   Output workspace containing calculated pattern or calculated background.
InputParameterWorkspace Input TableWorkspace Mandatory Input table workspace containing the parameters required by LeBail fit.
OutputParameterWorkspace Output TableWorkspace   Input table workspace containing the parameters required by LeBail fit.
InputHKLWorkspace Input TableWorkspace Mandatory Input table workspace containing the list of reflections (HKL).
OutputPeaksWorkspace Output TableWorkspace   Optional output table workspace containing all peaks’ peak parameters.
WorkspaceIndex Input number 0 Workspace index of the spectrum to fit by LeBail.
FitRegion Input dbl list   Region of data (TOF) for LeBail fit. Default is whole range.
Function Input string LeBailFit Functionality. Allowed values: [‘LeBailFit’, ‘Calculation’, ‘MonteCarlo’, ‘RefineBackground’]
PeakType Input string ThermalNeutronBk2BkExpConvPVoigt Peak profile type. Allowed values: [‘ThermalNeutronBk2BkExpConvPVoigt’, ‘NeutronBk2BkExpConvPVoigt’]
BackgroundType Input string Polynomial Background type. Allowed values: [‘Polynomial’, ‘Chebyshev’, ‘FullprofPolynomial’]
BackgroundParameters Input dbl list   Optional: enter a comma-separated list of background order parameters from order 0.
BackgroundParametersWorkspace InOut TableWorkspace   Optional table workspace containing the fit result for background.
PeakRadius Input number 5 Range (multiplier relative to FWHM) for a full peak.
PlotIndividualPeaks Input boolean False Option to output each individual peak in mode Calculation.
IndicationPeakHeight Input number 0 Heigh of peaks (reflections) if its calculated height is smaller than user-defined minimum.
UseInputPeakHeights Input boolean True For ‘Calculation’ mode only, use peak heights specified in ReflectionWorkspace. Otherwise, calcualte peaks’ heights.
Minimizer InOut string Levenberg-MarquardtMD The minimizer method applied to do the fit, default is Levenberg-Marquardt. Allowed values: [‘BFGS’, ‘Conjugate gradient (Fletcher-Reeves imp.)’, ‘Conjugate gradient (Polak-Ribiere imp.)’, ‘Damped GaussNewton’, ‘FABADA’, ‘Levenberg-Marquardt’, ‘Levenberg-MarquardtMD’, ‘Simplex’, ‘SteepestDescent’, ‘Trust Region’]
Damping Input number 1 Damping factor if minimizer is ‘Damped Gauss-Newton’
NumberMinimizeSteps Input number 100 Number of Monte Carlo random walk steps.
MCSetupWorkspace Input TableWorkspace   Name of table workspace containing parameters’ setup for Monte Carlo simualted annearling.
RandomSeed Input number 1 Random number seed.
AnnealingTemperature Input number 1 Temperature used Monte Carlo. Negative temperature is for simulated annealing.
UseAnnealing Input boolean True Allow annealing temperature adjusted automatically.
DrunkenWalk Input boolean False Flag to use drunken walk algorithm. Otherwise, random walk algorithm is used.
MinimumPeakHeight Input number 0.01 Minimum height of a peak to be counted during smoothing background by exponential smooth algorithm.
AllowDegeneratedPeaks Input boolean False Flag to allow degenerated peaks in input .hkl file. Otherwise, an exception will be thrown if this situation occurs.
ToleranceToImportPeak Input number Optional Tolerance in TOF to import peak from Bragg peaks list. If it specified, all peaks within Xmin-Tol and Xmax+Tol will be imported. It is used in the case that the geometry parameters are close to true values.

Description

This algorithm performs Le Bail Fit to powder diffraction data, and also supports pattern calculation. This algorithm will refine a specified set of the powder instrumental profile parameters with a previous refined background model.

Back to back exponential convoluted with pseudo-voigt

Here is the list of the peak profile function supported by this algorithm.

  • Thermal neutron back-to-back exponential convoluted with pseudo-voigt
    • geometry-related parameters: Dtt1, Dtt2, Zero
    • back-to-back exponential parameters: Alph0, Alph1, Beta0, Beta1
    • pseudo-voigt parameters: Sig0, Sig1, Sig2, Gam0, Gam1, Gam2

Thermal neutron back to back exponential convoluted with pseudo-voigt

Here is the list of the peak profile function supported by this algorithm.

  • Thermal neutron back-to-back exponential convoluted with pseudo-voigt
    • geometry-related parameters: Dtt1, Zero, Dtt1t, Dtt2t, Width, Tcross
    • back-to-back exponential parameters: Alph0, Alph1, Beta0, Beta1, Alph0t, Alph1t, Beta0t, Beta1t
    • pseudo-voigt parameters: Sig0, Sig1, Sig2, Gam0, Gam1, Gam2

Optimization

LeBailFit supports a tailored simulated annealing optimizer (using Monte Carlo random walk algorithm). In future, regular minimizes in GSL library might be supported.

Supported functionalities

* LeBailFit: fit profile parameters by Le bail algorithm; 
* Calculation: pattern calculation by Le bail algorithm; 
* MonteCarlo: fit profile parameters by Le bail algorithm with Monte Carlo random wal; 
* RefineBackground: refine background parameters

Source

C++ source: LeBailFit.cpp (last modified: 2017-11-30)

C++ header: LeBailFit.h (last modified: 2016-10-24)