Induction generator

[zolka]

Helo!
I’m working on an induction generator simulation. But I have a problem when it comes to the state-space equations. If I assume that the current and voltage are zero at the starting moment of the simulation, then the state-space equations will always give the result that voltage=0, current=0, no matter how many times it’s iterated. So, the current and voltage must have a very little positive value at start! The question is: How can I estimate or calculate those values? Or, where can I find induction generators with those given values?
If you can give me some advice, please don’t hesitate!
Thanx

 

[electricpete]

I assume you are starting the simulation at zero speed?

You should have some representation of the nonlinear characteristics of the iron built into your model? It should include a residual flux density Br. That is what gets things started.

What are the state variables in your model?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[waross]

Start with a little residual magnetism.
Or-
Start with a residual voltage of a few volts. Current will depend on the total circuit impedance.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[zolka]

Yes, I start at zero speed.
The state variables are: Rs, Ls, Rr, Lr, Lm, omega.

 

[electricpete]

Rs, Ls, Rr, Lr, and Lm do not sound like state variables to me. The state equation is x' = f(x,t) where x is a vector of state variables (should include currents and/or fluxes).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

More importantly, those parameters suggest a linear model.

You need a non-linear model. The value of rotor flux corresponding to 0 current in your non-linear model is what you need factored into your initial conditions.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[zolka]

Sorry, I mixed up. The state variables are the currents.

 

[electricpete]

My comments about the importance of residual magnetism are under the ASSUMPTION the generator will be started as a self-excited machine. (I'm not sure if there is any other way to start an induction generator, but I figured I'd better mention that)

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

If I get a chance this weekend I will do a state space solution. It seems pretty simple to me from your equivalent circuit with the complication that we use a non-linear characteristic instead of Lm.

There would be three state variables. One would be speed. The other two would be two of the three three complex currents I1, I2, Im. The third current is calculated from the other two. We need to assume a driver torque profile vs time. We need to assume a non-linear Vm vs Im characteristics (takes the place of Lm). We could assume an externally connected constant impedance Z. If external impedance is open circuited (generator is starting up under no-load), then we drop one of the current state variables; I1 = 0, I2 = IM.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

If external impedance is open circuited (generator is starting up under no-load), then we drop one of the current state variables; I1 = 0, I2 = IM.

Scratch that. We need at least a capacitor for externally connected impedance. (I am still assuming the generator is starting without any externally applied voltage excitation)


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rbulsara]

I am waiting for more knowledgables to chime in, plus I have no knowledge of simulation. But assuming any simulation will simulate reality, consider the following:

When all three (rpm, voltage and current) are simultaneously at zero not much happens to any electrical machine let alone the induction gens.

Induction generator by definition have no self excitation when starting from the rest. They are either started as induction motors (rarely) and engaged to a prime mover OR more typcially, brought to near synchronous speed by a prime mover and a breaker is closed in connecting the machine to a live power source (grid or synchronous generators).

They only start exporting power when run slightly above the sync speed while still connected to a "grid".

The modern wind turbines with induction machines, for example, uses a bi-directional static (thyristor) switch to connect a induction generator to the grid to minimize impact (inrush of reactive power).

There usually is no need for active sychronization as no voltage is generated by just spinning a induction generator (or very little).

IM getting self excited (momentarily) due to the caps, upon sudden loss of utility (while running) is not a starting condition.



Rafiq Bulsara

http://www.srengineersct.com

 

[rbulsara]

Or in other words, for a induction machine to act as a generator (exporting power) the speed has to be above synchronous speed and a "firm" live source of power at rated voltage and frequency has to be already present. At sub-synch speed the machine will act as a induction motor.


Induction machine is NOT a voltage generator by itself.

Rafiq Bulsara

http://www.srengineersct.com

 

[electricpete]

Yes absolutely the electrical frequency is above the rotational speed. The rotational speed is the state variable since it's derivative is known. The electrical frequency must be calculated from the rotational speed and the currents. It is not a straightforward calculation but that's how it must be done for a state space solution. You could not make electrical frequency a state variable because there is no equation for rate of change of electrical frequency.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Induction generator by definition have no self excitation when starting from the rest. They are either started as induction motors (rarely) and engaged to a prime mover OR more typcially, brought to near synchronous speed by a prime mover and a breaker is closed in connecting the machine to a live power source (grid or synchronous generators).

The generator can be started using capacitors for excitation. Residual magnetism gets it started. It does not require connection to the grid. That is the condition I intended to simulate since it was stated the machine was to be started from zero speed.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Yes absolutely the electrical frequency is above the rotational speed

Whoops. I had my motor hat on. The other way around of course.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rbulsara]

What do you mean by getting started? Rotations or production of power?




Rafiq Bulsara

http://www.srengineersct.com

 

[electricpete]

Producing power. A prime mover is required and induction generator with caps connected can start up and feed a load without being conntected to any external power source.

The simple system I had in mind to simulate has no controls - just choose values for parameters including capacitance and see what you get. The part mentioned about determing frequency from speed and current is a pretty challenging part. It is not explicity determined from the equivalent circuit since equivalent circuits impedance depends on frequency itself... but is implicity determined. Also it is not obvious how one would go about estimating he actual non-linear magnetic characterics for a given machine from nameplate data - first thought is just compute reluctance of series combination of some thickness airgap and some thickenss of iron in proportionas that resemble the physical motor. Make that shape pass through the nameplate operating point. All in all a little more challenging than I was thinking.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Sorry - I shouldn't have said it would match nameplate operating point - instead would intersect the linear system with slope Xm... the intersection occruing at the nameplate voltage.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[zolka]

Helo!
In my model, there are capacitors parallel with the stator windings. There is no external voltage or current applied to the stator or rotor either. Due to residual magnetism and the capacitors when the rotor starts rotating an initial flux, current, voltage appears. As the rotor picks up speed those values increase. This lasts until the rotor reaches it final speed, a little above synchronous speed. The self-excitation process is at no load. This is what I want to simulate. On the other hand, there are so many non-linearities that I doubt my model will give correct results. The state-space equations (and lot more) are described in the book: M. Godoy Simoes, Felix A. Farret : Renewable energy systems. This is what I primarily use.
Electircpete, you mentioned above that you will do a similar simulation (or equation), if you have any results or ideas I am looking forward to hearing from you!
Now I am searching for a machine (IG) to test my model!
If anyone has one, please let me know!

 

[rbulsara]

Any induction motor is a induction generator.

Rafiq Bulsara

http://www.srengineersct.com