PeakIntegration change

[Vickie Lynch]

For the IkeidaCarpenter fit of TOF data from time slice integrated data, use TotIntensity column of the table which does not have the background subtracted.

[Vickie Lynch]

Refs #4835 use TotIntensity from time slices for IkeidaCarpenter

Changeset: 80efd3fd3762ab3aa871d004f8ccd0f828ca813c

[Shelly Ren]

With Vickie's help, I ran IntegratePeaksMD(InputWorkspace=ws+'_MD2',PeakRadius='0.12',

BackgroundOuterRadius='0.18',BackgroundInnerRadius='0.15', PeaksWorkspace=ws+'_peaksLattice',OutputWorkspace=ws+'_peaksLattice') and also ran PeakIntegration (see the script) below. Then I compared the intensities from both runs.

When the IkedaCarpetnerTOF was turned off, the intensities from both runs were close, about 10% off. When the IkedaCarpetnerTOF was turned on, the intensities from both runs were quite different. As the PeakIntegration algorithm is improved, I am closing this ticket for the release, but will reopen it for further improvement in the next integration.

Script:

# Basic parameters for Triphylite Crystal #Name of the workspaces to create ws = "TOPAZ_3131" filename = ws+"_event.nxs" LoadEventNexus(Filename=filename,OutputWorkspace=ws,FilterByTofMin='3000',FilterByTofMax='16000')

# Load optimized DetCal file #LoadIsawDetCal(InputWorkspace=ws,Filename="/SNS/TOPAZ/shared/Spectra/TOPAZ_8Sept11.DetCal")

# Spherical Absorption and Lorentz Corrections AnvredCorrection(InputWorkspace=ws,OutputWorkspace=ws,LinearScatteringCoef="0.451",LinearAbsorptionCoef="0.993",Radius="0.14")

# Convert to Q space ConvertToDiffractionMDWorkspace(InputWorkspace=ws,OutputWorkspace=ws+'_MD2',LorentzCorrection='0',OutputDimensions='Q (lab frame)', SplitInto='2',SplitThreshold='150')

# Find peaks FindPeaksMD(InputWorkspace=ws+'_MD2',MaxPeaks='100',OutputWorkspace=ws+'_peaksLa ttice') # 3d integration to centroid peaks CentroidPeaksMD(InputWorkspace=ws+'_MD2',CoordinatesToUse='Q (lab frame)',

PeakRadius='0.12',PeaksWorkspace=ws+'_peaksLattice',OutputWorkspace=ws+'_peaksLattice')

# Find the UB matrix using the peaks and known lattice parameters FindUBUsingLatticeParameters(PeaksWorkspace=ws+'_peaksLattice',a='10.3522',b='6.0768',c='4.7276',alpha='90',beta='90',gamma='90', NumInitial='20', Tolerance='0.12')

# And index to HKL IndexPeaks(PeaksWorkspace=ws+'_peaksLattice', Tolerance='0.12')

outWS = "outWS" Rebin(InputWorkspace=ws,OutputWorkspace=outWS,Params='1000,-0.004,16000',PreserveEvents=True)

PeakIntegration(InputWorkspace=outWS, InPeaksWorkspace=ws+'_peaksLattice',OutPeaksWorkspace=ws+'_peaksLattice',IkedaCarpenterTOF=False)

[Vickie Lynch]

Refs #4835 fix function evaluation for IkeidaCarpenter

Changeset: 60ea7f8aad26c495223e2cb4736c4507a14645fa

[Vickie Lynch]

Refs #4835 results closer to IntegratePeaksMD

Changeset: b54f7876d3cdaffc249d40ce7b635cc75a052929

[Vickie Lynch]

Refs #4835 Larger intensity on windows

Changeset: 864c933fb5d197f97c1003fac8d62cedc9fd3a65

[Vickie Lynch]

Tests were added for IkeidaCarpenterTOF option. Results now are close to those of IntegratePeaksMD. Curvefitting changes removed the workspace needed for calculating the wavelength for the IkeidaCarpenter evaluation which was the main problem.

[Nick Draper]

Moved for inclusion in the 2.1.1 patch

[Vickie Lynch]

Refs #4835 fix function evaluation for IkeidaCarpenter

Changeset: d795d2ab85df945126b91742f55db2171254accf

[Vickie Lynch]

Refs #4835 need IntegratePeakTimeSlices changes for tests to pass

Changeset: ccd234ef13a1cb3f7976e8fdc52ec22703416bcd

[Vickie Lynch]

Refs #4835 new test for BivariateNormal

Changeset: 5e9a393e0f7dbdaeda737180cd674dc73e0b01f8

[Peter Peterson]

A random survey of peaks integrated with MD methods versus this are in decent agreement.

[Stuart Campbell]

This ticket has been transferred to github issue ​5682