Fathom QiQuac M5, Руководство по эксплуатации

Страница: 15 / 33

15
SDIQ Grading
The QQ will attempt to grade your measurement. For this to be possible both
channels must have made a measurement so that they can be compared. The
Grade thresholds are set in Setup > SDIQ Settings > Grade A Threshold % and
Grade B Threshold %. These are set to 7% and 15% by default. These are
meant to represent 2 sigma (95% confidence) intervals and represents thresh-
olds from BC RISC guidelines for Grade A and Grade B measurements. Grade
A is assigned if QUnc. for Ch0 and for Ch1Ch1 and the percent difference in Q
from the two channels (%DQ) are all less than Grade A Thresh%, likewise for
Grade B. Grade C is assigned if at least one of these three metrics is greater
than Grade B Thresh %.
In our example, the Sensor on the other bank of this trial resulted in a Q of
1.42 m³/s ±4% which is 8% different than this sensor and therefore a B Grade
was assigned with the suggestion to increase the mixing length.
SDIQ Summary File
The QQ produces a Summary File with an Average Q between Ch0 and Ch1Ch1.
The uncertainty associated with the Average Q is
(11)
This estimate assumes the maximum uncertainty from the 3 estimates should
be assigned to the measurement.
SDIQ Uncertainty
Summary
Unlike the Infinite Improbability Drive on the Heart of Gold (Adams, D. 1979)
there is nothing to be gained from large uncertainty in an SDIQ. You are trying
to achieve a Grade A measurement. There are various best practices to achieve
this. The QQ tries to capture the significant features of the measurement quan-
titatively in order to assign a Grade and suggest ways to improve your measure-
ment if a Grade A is not achieved. You can almost always decrease the QUnc. on
both channels by injecting more salt to overcome any natural background EC.T
variations, although more salt will never overcome a problem with incomplete
mixing. In that case, increasing the mixing length with usually overcome this
problem. Be careful to avoid local inflows, recirculating pools and eddies, and
aeration (bubbles). These will all increase the QUnc and the DQ. It’s good prac-
tice to make two injections to determine the variability in your method, assum-
ing the stage level is relatively constant. Above all, “Don’t Panic.”
C H A P T E R 2 : T H E ORY