Simulink: Synchronous Generator plus controlled voltage source - simulation is very slow

Hi there guys

So, I'm trying to simulate a small three-phase isolated power system, which consists of a synchrounous generator (with full governer+turbine and excitation blocks), a transformer, a certain load and a three-phase controlled voltage source modelled with 3 single-phase "controlled voltage source" blocks, with a Y configuration to a common ground.

I want to simulate the outage of the synchronous machine at a given time, such that only the 3-phase controlled votlage source remains in service, supplying the load. This is beeing done in the "continuous" mode. The problem that I have is related to the simulation time: to simulate about 30 seconds it takes a lot, I mean a LOT of time.

I understand that modeling the synchronous machine eletromechanical dynamic behaviour isnt quick but, I'd like it to be faster.

I previously tried the exact same system in "phasor" mode simulation and it proceeds faster, but the frequency deviation and all the other measurable variables, such as voltage and ative power appear with very high oscillation in the simulated time scale.

Is the 3-phase controlled voltage source in the format that I've constructed it, compatible with the "phasor" simulation?

Is there any way to speed up the simulation time in the "continuous" mode?

Thank you very much, André

Answer 1

[Sebastian Castro]

What solver are you using? For SimPowerSystems models, the "ode23tb" solver is recommended, and simulation can be pretty slow with others. Look at this link:

http://www.mathworks.com/help/physmod/sps/powersys/ug/simulating-with-continuous-integration-algorithms.html

- Sebastian

[André Silva]

Hi there Sebastian!

Thank you very much for the reply! Changing the solver to "ode23tb" indeed fastened the simulation, but didnt solve the main problema. So, I decided to spend some time analysing the control loop for the 3-phase controlled voltage source and ended up realizing that the main problem was related to the P-f droop I had implemented. It's working just fine right now.

Thanks again, best regards

André

[Jubeyer Rahman]

Hi Andre,

I am working on a project which needs the P-f droop implemented. Can I have your model? My email Id: [e-mail] Thanks.