Rockwell Automation FactoryTalk Historian SE ProcessBook 3.2 User Guide - Manual (Page 178)

●
●
●
●
●
FactoryTalk Historian ProcessBook User Guide
160
1. Plot the temperature of Boiler1 (the B1Temp tag) on the X-axis of an
XYPlot and the temperatures of the other three boilers (B2Temp,
B3Temp and B4Temp) on the Y-axis.
2. Use the same time range for all four tags.
3. Select a single scale so that the Boilers 2, 3 and 4 are compared
directly against Boiler1.
4. Configure the plot to show the correlation co-efficient for each of the
boilers.
By viewing how far from the regression line each of the boilers falls, you
can determine how closely their performance matches and which boilers
you should continue to adjust. When all three boilers have an acceptable
correlation coefficient, you know the work is complete.
Example 3: Lab Comparisons
An engineer wants to compare lab results from his lab to those of another
lab for the same sample of material.
The plant's quality assurance lab has instituted some new testing
procedures. In order to gauge the validity of the new testing procedures, an
engineer wants to compare results from the new process against data from
an outside lab that is known to have valid testing procedures.
1. Put the results from the in-house lab into a data set.
2. Place the results for the same test from the other lab into another data
set.
3. Selects each data set as a tag in the plot, and select which one to be the
X-axis.
4. Choose Recorded as the retrieval mechanism for each data set so that
the results are not skewed by minor time stamp differences.
If the XYPlot shows the results from the new process are well correlated
with the outside lab's data, the new test procedure is validated.