# Low Power Inverter Imports Energy?

A lab-sized microgrid (island) has been set up to test some equipment. An inverter+battery is being used as the source, with the inverter operating at very low power compared to its rating. This makes the system PF low and capacitive.

What's confusing me is looking at our power and energy measurements from the inverter's output. With positive indicating energy/power import, and negative indicating export, here is a graph of power (top) and energy (bottom):

Here, power is consistently negative, as one would expect from an inverter's output. But energy oscillates between negative and positive at a regular interval. The magnitude of the negative is twice that of the positive, so overall it's reading energy exports, but why would it ever be importing?

My one guess has to do with how the meter in question measures energy, which is as an accumulated field (ie it just has 2 counters going up, one for imports, one for exports). If there was a teeny amount of energy imported at certain points of the inverter's switching cycles, even too low to be measured by the instantaneous power readings, it would eventually accumulate to an amount that would be read by the energy field.

That's just a hypothesis though, would appreciate someone who could explain what's going on here in more detail.

EDIT: Worth noting is that the positive spikes are correct, in that if they're disregarded as erroneous, energy and power do not match up at all (energy accumulates about twice as fast as it should), however if those positive spikes are added to the negatives, energy accumulates at the rate expected from a 60 something watt load.

• How are you measuring? Can you show a block diagram of your existing setup? – winny Nov 12 '19 at 7:52

Alright, so this isn't really an EE anomaly as much as a metering issue. What's happened is the meter we're using has two accumulated energy measurement fields. The first is Total kWh, which measures all energy in either direction flowing through the meter. The second is Reverse kWh, measuring the energy flowing out from the load (production). The meter doesn't measure, but only calculates Forward kWh, with the formula Forward = Total - Reverse. That alone seems weird to me. Why not just measure forward and reverse? Who cares about total?
But this problem appears because, due to rounding, sometimes Reverse increments w/o also incrementing Total. This causes the calculated Forward to decrement, screwing up the readings. Eventually, Total will increment w/o incrementing Reverse, "correcting" the problem, but also causing Forward to increment incorrectly.