PT Accuracy Class 3P & 0.2S

Besides the possibly unknown meanings of your acronyms and units, something sounds strange here. Do you really have a Three Phase Power Transformer that converts from 132kV directly to 110V at a power level of only 30VA?

IEC defines metering accuracy for a protection CT core. For example, 5P core will have metering accuracy class 1 in load current range. I couldn't find similar notes for protection VTs.

I have been specifying dual accuracy class for VTs like 0.5/3P and vendors do not have a problem supplying the same.

In your case, you may test the protection VT that you may already have at site, for accuracy between 80 & 120% of the rated voltage (and burden between 25 & 100% of rated). Or else, approach the vendor for advice, giving VT name plate details.

thanks for your valuable suggetions.

http://www.rsisolsec.com/en/accuracy-of-current-transformers.html

The coil for relay protect not use for energy accounting.

You are trying to calculate the difference between the readings obtained from a metering-class PT on the utility side, and an instrumentation-class PT feeding the Customer Revenue Panel. You will need to obtain the calibration curves for both, as well as the burdens of their respective circuits, then calculate how each will react under changing primary voltages/loadings, and compare the results.

The alternative would be to energize the bus with no load, inject varying current steps into the CT secondary circuit, and record the readings on both kW/kWh meters.

+/- 3% between 5% and 100% of rated primary voltage and between 25% and 100% of rated burden (at 0.8pf) connected to the secondary.