Chapter 5 _______________________________________________________ Functional Description
VAISALA _______________________________________________________________________ 105
The software normalizes the measured signal and then estimates the
precipitation intensity using an empirical algorithm. The estimate is
proportional to the amount of water on the DRD12 surfaces. The
signal normalization is calibrated at the factory with dry and
completely wet surfaces.
When the heating of the sensor elements is switched off, the sensor
elements become very sensitive to ambient moisture. If the relative
humidity of the air is more than about 70 % , the surfaces produce a
measurable signal even with no precipitation. This is used as an
estimated humidity measurement in the FD12P. The estimated
humidity is used to separate between haze and mist.
As the FD12P can detect the beginning of precipitation from the
optical signal, the DRD12 heating can be turned off when there is no
precipitation. The software disconnects the heating after one hour
from the previous precipitation detection. When precipitation or
moisture is detected again, the heating is turned on for at least one
hour.
Algorithm Description
Visibility
The optical signal analysis calculates the difference of the
measurement signal and offset averages for the visibility algorithm.
The difference value (frequency) is given as a parameter to a
calibrated transfer function. The transfer function converts frequency
into visibility (MOR). A signal of one hertz corresponds to a visibility
of about 4500 m and 100 Hz about 150 m. The exact form of this
transfer function has been defined using an accurate transmissometer
(Vaisala MITRAS) as a reference.
The instantaneous (15 s) visibility values are averaged to get one- and
ten-minute average output values. The averages are calculated from
extinction coefficient values to better emulate human observations.
The extinction coefficient (
σ
) is defined as follows:
σ
(1/km) = 3000/MOR (m)
If the signal level is less than 0.23 Hz (visibility over 15 km), the
FD12P uses extra filtering in the one-minute average MOR value. The
filter function is the following:
