Line Drop Compensation - On-Line Tap Changer 1

[Adam1980]

Dear members,

i am trying to simulate the calculations done by the line drop compensator of a tap changer in order to estimate the voltage at the end of a 110 kV 120 km line based on local measurements of voltage, current and current phase angle.

The supplier of the tap changer insists that the remote voltage will be estimated correctly once the LDC parameters are defined correctly and an offset to the voltage set-point of the tap changer to compensate the no-load voltage rise across the cable is applied.

However based on my simulations and when i am using the measured current and current phase angle the remote voltage will not be estimated correctly since the charging current of the cable is not defined in the LDC equations that I found in literature. This error will in crease with decreased loading of the cable.

My main question is: Do the LDC calculation or measurement circuit have certain methods of removing the charging current from the local measured current and phase angle?

I would appreciate any pointers.
Thank you.
Adham

 

[dpc]

I've never seen an LDC that factors in the capacitance of the line. Normally these are used on distribution substations where the inductive reactance swamps any line capacitance, and feeder distances are much shorter than 120 km. I think you are expecting a much higher level of accuracy than is normally considered necessary.

But the shunt charging current is not a function of load. So its impact should be able to be handled by the voltage setpoint?

You could probably roll your own LDC using a microprocessor relay that has both voltage and current inputs. I have seen tap changer voltage setpoint control done using only an SEL-351 relay. Or a small PLC, perhaps.

 

[Adam1980]

Thank you dpc.

The voltage rise caused by the capacitance is independent of the loading level however the local current magnitude and angle which i am using for the R and X drop will be affected by this capacitance. correct?

 

[magoo2]

The standard LDC just has an R and X setting, so it does not consider any capacitance effects.

You should be able to calculate the voltage drop under a range of conditions for your line. You can then do the same for the line drop compensator. Then you can compare the two results.

You mentioned both line and cable in your posts. If it is underground cable, then the capacitance is quite significant. If it is a line, it is much less significant but still ought to be considered.

 

[slavag]

What dpc and magoo2 said.
only two parameters X and R (Ur and Xr).
You can use another parameters for cable capacitance compensation: U set points, bandwidth,etc..

 

[Adam1980]

Thank you for the replies.

It is actually a subsea cable which has a 350 A charging current.
is there any reference of how the calculations are made for the LDC i just want to make sure that i am using the correct procedure.

Perhaps i am expecting much accuracy as was pointed out before but to my understanding from the supplier of the tap changer the LDC will be able to estimate the remote voltage accurately.

I am wondering if some LDC have the option of including the charging current in the algorithm. This value if subtracted from the reactive part of the local current measurement, the remote voltage could be estimated correctly.

Otherwise i can not see how an LDC calculation could make correct estimation if the current measured is 300 A (including charging current) while the current which will actually lead to the drop is only 180 A.

Thank you
Adham

 

[slavag]

What is nominal current of cable?
What is X and R of cable?

 

[slavag]

BTW, possible think in your case about another interesting solutions.
I'm assume, you have some communication link between ends of the cable and of course voltage metering on the second send of the cable.
Possible use a newer devices and send voltage analogue information to tap changer control by IEC61850.

 

[Adam1980]

R is 13.15
X is 15.76

no-load current 346A and full load current is 352A measured locally at the secondaries of the transformer. voltage rise is 4.3%.

what is meant by the nominal current of the cable? ampacity?

Thank you.

 

[slavag]

Interesting, very interesting situation.
I meant nominal object current, in your case 352A.
What is tap data? in voltage percents?

 

[slavag]

Could you please say, what is a tap changer control device?
I think a best option build some additional logic, but for this need information from second side of cable.

What is no load situation? Circuit Breaker on the second side is opened?

 

[Adam1980]

+- 10 taps each 1,5% on the secondary side of a three winding transformer. I am trying to control my remote remote voltage to 2% from the nominal.

The voltage to be controlled is on the secondaries of the transformer.

I believe it is very interesting situation since I do not have previous experience with LDC as well. However when I usually think of an acceptable error that would be in 1% range but here it is going from 10% additional to 10% less estimation which is no more acceptable I believe.

I tried to find references on how the LDC is calculated from VR manufacturers so I reached the conclusion that they are either too complicated and smart to post or too straightforward. I think it is the latter.

Thank you.

 

[Adam1980]

To tell you the whole situation.

I am controlling the voltage on an offshore platform from an onshore substation. usually i have a fiber optic cable getting measurements from the remote location. however when this line is lost i want to make sure that my local tap controller can still control the remote voltage to a tight band using only local measurements and LDC estimation.

The no load situation could be only cable connected or very small offshore load. i will list next the load profile:

Load (MW) Current measurement at the LDC (A)
0 346
10 321
20 305
30 299
40 307
50 329
57 352

assuming my offshore load is always with 0.9 power factor.

Thank you.

 

[slavag]

Thanks for information.
I'm not have experience with LDC too, but it's not so important, theory is theory.
Could you please add to your table voltage measurement in percent of nominal?

really interesting situation!

 

[slavag]

Adam1980.
What is tap changer control device?
It's very important.

 

[Adam1980]

In the same sequence the measured remote voltage in % is:
104.36 <---- 4.36 is the constant voltage rise.
102.59
100.75
98.83
94.71
93.16

controller is TAPCON260 i believe it is the same as others.

thank you.

 

[Adam1980]

All values are made assuming a 100% local voltage and nominal tap position.

i will add also the current phase angle (cosphi) to ease any calculations you are making, same sequence as before:
-0.02
-0.18
-0.36
-0.55
-0.72
-0.85
-0.91

 

[slavag]

I'm not see any option build suitable logic in Tapcon260.
You need something with cosPhi control, both option of compensation in the Tapcon260 are not included such option ( according to my understanding).
From you table we can see:
"low" value of cosPfi- "overvoltage" situation, need block LDC and use a simple Auto regulation.
"high" or nominal value of cosPhi- "undervoltage" situation and need use LDC compensation.
near to cosPhi=0.5 voltage is equal to 100%.

We can see, that all measurements according to theory.

Problem of standard tap changer control is current measurement, for the devise is not important, active or reactive current.

I try found in manual option of change settings in Tapcon260 by any binary/digital input.

 

[Adam1980]

Thank you slav.
just to make sure that i am getting my calculations correct.
By the LDC calculations i will get those voltages in % at the remote node if i used same sequence of input data i included earlier.
91.85
93.7
95.63
97.65
99.76
101.98
103.61

which shows that even if i used a constant offset for the setpoint of 4.36 which is the initial voltage rise i will not reach the actual remote voltage.

when in my LDC calculations i removed the no-load charging current from the measured reactive current and then calculated a corrected current magnitude and phase angle.The LDC calculations leads to the following remote voltage results in %.

For minimum loading:
actual current measured: 346 A
actual current phase: -0.02
actual remote voltage: 104.36%
calculated remote voltage with normal LDC: 91.85% + 4.36% = 96.21%
calculated remote voltage with LDC deducting the charging current:
100% + 4.36% = 104.36%

For maximum loading:
actual current measured: 352 A
actual current phase: -0.91
actual remote voltage: 93.16%
calculated remote voltage with normal LDC: 103.61% + 4.36% = 107.91%
calculated remote voltage with LDC deducting the charging current:
88.81% + 4.36% = 93.15%

any ideas if this is usually implemented in an LDC?
slav maybe we should exchange emails to check on our calculations.

Thank you

 

[magoo2]

Can you show the impedances involved including the capacitance? This should include the transformer impedance. I think you said this was a 3 winder.

I am not familiar with your specific control but I know we controlled a cable-fed transformer from the other end using a bias voltage. It was at 34.5 kV but the application sounds very similar.