simulation of induction generator

Discussion in 'The Projects Forum' started by simulation, May 17, 2009.

  1. simulation

    Thread Starter New Member

    May 17, 2009
    1
    0
    i am doing small project on wind power generation.
    so i have to model the induction generator .i dont know how to do the simulation of induction generator
    can anybody give me the suggestion plz?/
     
  2. steveb

    Senior Member

    Jul 3, 2008
    2,433
    469
    Can you give us a background on your level of math skills, and what tools you will use to implement a simulation of the induction motor?
     
  3. rvh002@gmail.com

    Active Member

    May 15, 2009
    118
    2
    Indeed? There are bits of reference in the AC section of the reference book at the top of the page that show how to turn a motor into a generator. Go to AC and the to section 13 and to Tesla motors for a start .
    Regards
     
  4. Mike Mandaville

    Active Member

    May 27, 2009
    81
    1
    Here is the original generator:

    [​IMG]

    Hippolyte Pixii called it a magneto-electric machine.
     
  5. Mike Mandaville

    Active Member

    May 27, 2009
    81
    1
    We have not heard from Simulation again, but this topic interests me. It seems to me that the first thing which would be required is an estimate of the amount of wind which could be expected to blow at the location where the generator would be set up.

    I personally think that an Alaskan husky would be much more reliable. Such an animal can lope, trot and walk twelve hours a day, in harness, and pulling a sled. Take one out of harness, put him in a circular treadnill, add a magneto, and you will have some reliable power.

    Here is a hamster-sized version:

    http://otherpower.com/hamster.html
     
    Last edited: Jun 10, 2009
  6. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Damn I want one! That would make a real nice cellphone charger for hippies. It's even got a cute little commutator!
    ;)
     
    dyhia likes this.
  7. Mike Mandaville

    Active Member

    May 27, 2009
    81
    1
    For alternating current, I think that the Pixii design is the best. For direct current, the Clarke design is simpler:

    [​IMG]
     
  8. Mike Mandaville

    Active Member

    May 27, 2009
    81
    1
    And, if you don't want to turn the crank yourself, you can delegate that responsibility to your dog:

    [​IMG]
     
  9. alioshak

    Member

    Jul 31, 2009
    10
    0
    I am trying to developing an induction generator model,too
    by using state space
    but I am confused about inputs, I have reseached a lot, and every paper that I ´ve read use the Voltage values (dq axis) and the mecanical torque as inputs
    I agree with toruqe, but why the Voltage are inputs too?
    would anybody help me?
     
  10. steveb

    Senior Member

    Jul 3, 2008
    2,433
    469
     
  11. alioshak

    Member

    Jul 31, 2009
    10
    0
    thanks to answer me Steveb

    but my idea was that the fluxes (ψ) could be the states, so both of them rotor and stator currents would be the outputs (becuase I need to study a generator couple to the grid, so the generator has to provide current and voltage) and as I said torque would be the input
    but, to calculate the currents values I need the Voltage (Vds and Vqs becuase Vdr and Vqr are zero becuase I am studying a squirrel cage generator) , could I use Voltage sources as input in order to simulate (in matlab/simulink)? but what would be my justification?

    I hope you could answer me
    thank you very much
    see you

    Alioshak
     
  12. alioshak

    Member

    Jul 31, 2009
    10
    0
    thanks to answer me Steveb

    but my idea was that the fluxes(ψ) could be the states, so both of them rotor and stator currents would be the outputs (becuase I need to study a generator couple to the grid, so the generator has to provide current and voltage) and as I said torque would be the input
    but, to calculate the currents values I need the Voltage (Vds and Vqs becuase Vdr and Vqr are zero becuase I am studying a squirrel cage generator) , could I use Voltage sources as input in order to simulate (in matlab/simulink)? but what would be my justification?

    I hope you could answer me
    thank you very much
    see you

    Alioshak
     
  13. steveb

    Senior Member

    Jul 3, 2008
    2,433
    469
    You can use fluxes as states instead of the currents, but the same structure results. Basically, flux and current are related thru inductance and voltage is the time rate of change flux. Either way, current is an output from the statespace model and voltage is the input.

    The idea of voltage being an input is really confusing at first, isn't it? It seems crazy because the generator is providing both current and voltage as outputs into some kind of load. However, the statespace formulation considers the voltage as an input, which is just a mathematical thing.

    The basic idea here is that the output current from the generator goes into the load. The load model take the current as an input and generates a voltage response as the output. Then generator then takes the voltage from the load as an input. In reality what is happening is that the generator and load are interacting and are coupled.

    You could make a model in which the voltage is the output and the current is the input, but then it would not be a state-space model. However, state-space models are well behaved when modeled in Simulink, so people often use them when they can. As I mentioned in the previous post, if your load is a perfect constant current load, there is a conflict because you would like that constant current load to be a source output which drives the current input of the generator, and then let the generator produce the voltage response as an output at whatever value it needs to be.

    I'm assuming this is not your case because you mentioned connecting to a grid, but if you do want to drive a constant current load, the solution is to make a non-ideal model for the constant current load, and hope that the numerical solution converges. This whole confusing situation is a result of trying to make a load model separate form the generator model, when in fact they are one coupled system.

    You have the option of trying to generate a larger state-space model which includes the generator and load together. This is a preferred method when it is possible, particularly if you have a nonlinear state-space model. I've found that linear state-space models are well behaved when they are broken into subsystems with inputs and outputs, but you need to be much more careful with nonlinear state-space models.
     
  14. alioshak

    Member

    Jul 31, 2009
    10
    0
    Thank you very mach
    It was really usefull to me
    see you
    alioshak
     
  15. onlyvinod56

    Active Member

    Oct 14, 2008
    362
    1
    hi alioshak,
    may i know why u r interested in state space modelling.....
    u can use simulink where u can easily study the output frequency of the stator with different speeds of the shaft at different slip frequencies of the rotor.

    u can make the ind.gen to work like a var compensator.
    in simulink, slip[ ring ind.motor models are available.
    just couple a dc motor to the slip ring ind.motor.
    feed a three phase ac to the rotor thru slip rings.
    observe the stator output.
    run the dc motor at different speeds and observe the output frequency.

    respond ur output here.dont forget.
     
Loading...