Eh? 0-d-q system 1

[pavement]

Can anybody introduce myself to the joys of Parks Transformation, given that i have induction machine parmeters in both physical and per unit sense. This is first time i have encountered it ,as a new package has been brought in.
How does the physical parameters relate to the matrices given in various texts, absoulutely bamboozled?, as no texts gives examples on how it is carried out.

Any help from the experts out there would be very appreciated.

 

[DougMSOE]

Park's or Blondel's transform are both methods by which one can change phase values to D-Q-0 and the reverse. If you start with the line to neutral phase impedance, the transform matrix will get the D-Q-0 values. I have always prefered Blondel's transform matrix but both will get the same values. Further both can be used with per unit or with real values. Also take a look at symetrical components. These are basically the static form of Blondel's transform.

 

[pavement]

Sorry for the ignorance, just that i have Ra,Xa quantities etc, but the package wants the rotor and stator impedances in the transformed quantities, i have the relevant texts but still have no idea how to proceed?

 

[rsherry]

You say that you have induction machine parameters - it is unusual (but not impossible) to adopt a dq0 analysis of an induction machine (even a wound rotor induction machine) for 2 reasons :
1) the assumption of synchronous rotation of rotor fluxes is not valid with induction machines
2) the power of an induction machine is usually defined as a function of the slip (from the torque-speed curve) and not from derivation of the rotor angle (delta) relative to the stator - i.e the position of the d-axis.
Are you considering motor or generator studies ?

 

[pavement]

Its for fault level studies,ive tried 'manually' tuning the machines to the correct outage which is fine, but when they are caught up in the fault scenario they dont behave as desired i.e the starting current characteristics dont die away and cause the currents drawn to increase to a higher steady state in two of the machines.Using ATP which specifies the following, when using the induction machine models....
Stator page
Specify resistance and inductance in Park transformed
quantities (d- q- and 0- system).
NB! All inductances in H or pu.
Rotor page
The total number of coils are listed and
given data on the Rotor page. First the the
d-axis coils then the q-axis coils.
Specify resistance and inductance for each coil.
All the coils are short circuited.
NB! All inductances in H or pu.

Due to the large sizes of the motors ( highest 13.1MW) involved they should play a significant factor in the fault level dynamics.Im just totally stuck on how to proceed.

Thanks for getting back to me.

 

[ijl]

see

http://pp.kpnet.fi/ijl/prog.htm

Although the purpose of the program is a bit different, it might give an idea, how the ordinary parameters should be used in ATP or ATPdraw.

 

[pavement]

ijl, thanks for that, but i was just wondering do you have any particuliar experience of ATP, i.e. past examples of machine data implemented in ATP always make things more acceptable regarding results.

Cheers!

 

[jbartos]

Suggestion: Visit

http://www.p-e-s.spb.ru/downloads/docs/SV_PWM.pdf

etc. for more info

R.H. Park introduced a new theory in 1920s for electric machine analysis to convert three-phase machine to two-phase machine using resolution into d-q axis plane.
d-q variables are associated with fictitious machine rotating windings with rotor at synchronous speed.

 

[ijl]

Are you using ATP only, or with ATPdraw ?
Can you list the data for one motor, to start with?

 

[pavement]

Using it with ATPDraw, got the Rule book but i think the rule is that it doesnt make any sense.What im looking for is an 1.315MW DOL rated machine, which ive got the following parameters Ra=0.265004, La=0.013075, Lm=0.331087, R1=0.463757 while L2 and R2=0.

Does this help?

Many Thanks

 

[pavement]

Oops, all the above values are Henrys with regards to the L's.

 

[ijl]

Are these per unit values, or absolute ?
(my guess is per unit)
Voltage ?

 

[thomyorke]

Sorry, absoulute physical units and rated at 13.8kV for the DOL connection.

 

[pavement]

Sorry, absoulute physical units and rated at 13.8kV for the DOL connection.

 

[ijl]

Well, that was not so easy! (I am not a big expert, but I have to continue, because I got involved.)The notation of your data is unconventional. I have to make some guesses. It seems reasonable to assume that:

Ra = armature resistance
La = armature (leakage) inductance
Lm = magnetization inductance

But R1, R2, L2 ?

R1 might be a starting resistance.
(Please correct me, if necessary)

R2, L2 cannot be rotor resistance and inductance, because they are zero. But because they are zero, they can be neglected (!?).

But that means that the rotor data is not given. When the rotor data is not given, the conventional guess is that the rotor resistance and inductance (as seen from the stator side) are the same as the stator resistance and inductance.

Rr' = Ra and Lr' = La

The instructions for the induction motor in ATP and ATPdraw are confusing. I have found the following to work well with the universal motor 3 model (UM3):

General:
stator coupling Y
Pole pairs 1
Rotor coils d:1 q:1
Magnet:
LMUD = LMUQ = Lm (put the magnetization inductance here)
Stator:
R(0) = 0
R(d) = R(q) = Ra ( put the stator resistance here)
L(0) = 0
L(d) = L(q) = La (put the stator inductance here, in H)
Rotor:
Use same values as for the stator

That is, do not care about the Park's transformation !

By the way, the induction motor is a tricky component in ATP. There must be some component between a source and the induction motor. A switch or a small impedance can be used.

If there are several motors in the same network, there must be a piece of transmission line connecting the motors to the network. The delay of the line must be longer than the time step, so that the line (in a way) isolates the motors from the network.

You can download a simple ATPdraw example of a network with an indcution motor at

http://pp.kpnet.fi/ijl/prog.htm

Your data has been used in the model, but note, the rotating speed is constant.

 

[pavement]

Well, if still to try that, but assuming youve already tried it yourself all i can say thank you and may Santa be good to you this year ijl.

Thanks for your no doubt valuable time.

Michael

 

[pavement]

Ijl, i've tried putting in those values, with the default slip of 0.53%, but outage is higher than expected. Can always fine tune, but thing that concerns me is that, when you apply a short circuit within the system, the quick asmmymmetrical characteristic you expect is nowhere to be found.Did you implement it in ATP, because it may be different versions having a say, if you did, could you be kind enough to send me the schematic to the following michaelmarkey@o2.co.uk so i could have a look to see where im going wrong. I already used switches for breakers via the HV bus, and my delta T is relatively small enough to avoid numerical instability??

Many Thanks

 

[Grollor33]

Pavement, I know this is an old issue for you but i'm having some of the same difficulties...did you ever get this resolved?

 

[ijl]

If I recall right, the issue was resolved. Wasn't it so, pavement? (I feel obliged and justified to answer, because I was involved . Jweav33, if your problem is purely ATP-related, it might be more fruitful to discuss it in the ATP-forum. See

http://www.emtp.org/

for more details.