Territory Stories

Electronic Data Collection and Analysis System

Details:

Title

Electronic Data Collection and Analysis System

Creator

Yin Foo, Des; Foley, Margaret

Collection

E-Publications; PublicationNT; E-Books; Report ; 39/1992

Date

1992-08-01

Description

Made available via the Publications (Legal Deposit) Act 2004 (NT).

Notes

Date:1992-08

Language

English

Publisher name

Power and Water Authority

Place of publication

Darwin

Series

Report ; 39/1992

File type

application/pdf.

Copyright owner

Check within Publication or with content Publisher.

Parent handle

https://hdl.handle.net/10070/229024

Citation address

https://hdl.handle.net/10070/673447

Page content

Technical Report WRD92039 Viewed at 15:07:17 on 29/07/2010 Page 172 of 192. D3. CALIBRATING LOGGED DRAWDOWNS BY MANUAL READINGS n DATt-'IANn IS "Calibrate" facility allows one set of data to bE' calibrated by another. the slope (s) and offset (0) through a plot of the set by The calibration parameters are of the best fit straight line which the calibration is done (the "true" set: (T}) against the set to be calibrated ({C}). I.e., the original values of {C} are multiplied by s and have o added to them, where sand 0 are from the best straight line through a {T} .v. {e} plot. Note that where there is no value of {C} at a time for which there lS a value of {T}, a value is automatically interpolated for {C}. This can lead to a a linear interpolation with respect to accurate where the interval between Therefore, some points may have to calibrating sea {T}. meaningless result as time is usually only readings is short. be deleted from the Before tl":,e calibration is completed, the {T} .v. {C} plot, and the s,o and r2 values are shown on screen. At this stage the plot can be inspected and any points which lie significantly off the best straight line can be noted. If there .He only a fe~l points significantly off the best straight line, the calibration can be aborted, and both data sets examined ']Iith a view to deleting those {T} values causing the anomalies. If there are many anomalous points or if the r2 value is lower than, fcr example, 0.980, serious consideration must be given to the vdlidity of the calibration.