Digital power meters, all measure real and reactive power.
The entire table of numbers is sent up the line. The digital meter
does not have the authority to meddle with the information it
collects. It just sends it.
It is up to the billing design, to show you all numeric fields,
or only show you the fields they bill on.
On a digital power meter, there are two sigma-delta converters
running at 500KHz or so. They are ADCs, and give you a string
of digital values. One is for Volts. One is for Amps. Using
a dedicated hardware processor, there are a standard set of
equations used for working out the quantities. If something
needs to be integrated, they integrate the area under the curve.
The high sampling rate (500KHz compared to 50Hz), ensures the
math is of good quality. There are regulations to be met on
accuracy, so that a digital power meter is every bit as good
as that old rotating platter thingy. As far as I know, it's
good to around 1% or so. Even if the current flow waveform is
"ugly", the meter properly measures that. Your digital multimeter
on the other hand, does not measure that waveform properly.
Your hand held multimeter samples at 1Hz or 2Hz or so,
whereas the power meter runs at 500KHz.
*******
As my power systems professor would quip, the "reactive power
sloshes back and forth between the consumer and the power plant".
All current flows, result in resistive heating and losses.
The penalty for a consumer, having too high a reactive component,
is a measure of the need to beef up components (transformers, wires)
to handle the extra current flow. Only the "real" component is
typically billed for on consumers, while industrial power users
are billed on reactive powers, so that the industrial user will
consider installing a capacitor bank, to reduce the magnitude of
the reactive power.
Some industries are extremely reactive on power, and so the
management know they need to correct their power factor, before
they even connect to the grid. It's not a "surprise on our
monthly bill, and geez, we need to fix this". They already know
by design it needs to be fixed.
*******
As for those power meters, someone actually found an error in
those equations. Too funny. You may find a reference to that,
if investigating "how a digital power meter works". The error
was not something that affected billables in a significant way.
And it happened quite a while ago.
One of the reasons for my interest in the topic, is my
hand held meters giving me bullshit readings (obviously
wrong values) and wondering what was going on. For example,
some computer here, it was soft off, and one of my meters
was indicating it was "using 100W". Which is absurd. A Kill-O-Watt
gives numbers which match conventional knowledge on the topic.
You can see the real component "W" and the reactive version "VA"
on the meter, and can see just how bad the power factor is on
the computer when it is soft-off.
My computer is using 49W, 91VA, PF=0.53, and on my power
bill I pay for 49W. No, the power supply is not modern, it
is not an 80+, and I could run off and change it over
to the spare I've got. But the power supply is compatible
with poor-quality UPS supplies (stepped "sine").
Paul