simpowersysteme discret simulation stability issue

I want to analyse the transient stability of a simple system over a long period of time. I do this because my ultimate goal is to run the simulation with real-time simulator for education purpose.

My model is really simple, one PV bus synchronous generator connected to a infinite swing bus via a bunch of inductance and resistance in serie and parallel. I simulate a fault at 0.5 sec and since i am simulating dampingless generator i expect my system to oscillate for the infinity

But.. When i run my model in phasor mode it run well and it do not diverge. But when i run it in discret mode with 20 us (as it will be on my real time simulator) the mecanical angle after the fault always increase until it reach instability.

btw i add up some extra load to help numerical stability as suggested in the help panel of the synchronous generator.

Answer 1

[Yahya Zakaria mohamed]

Nothing related to numerical solvers at all. I opened your model and found You have to simulated faults in your network.

1. The first at bus B1 at T=5000 and You set the switching on Time( T=5000) and didn't add the switching off Time(time to clear fault)
2. The second fault at bus B3 and You set the fault time at T=0.5 for six cycles (not practical the machine out of synchronize).

• The mechanical angle of the rotor is increasing for ever if the system is stable or not because the angle you seek for is the electrical angle between two buses (E and Va or V_B1 and V_B2).

I will consider the second which at T=0.5 and cleared after six cycles.

• You used fixed input mechanical power and a fixed input field voltage for the excitation and both are not practical and you won't stabilize the machine.
• We use speed controller for the Pm and digital excitation controller for the field current and both stabilize the machine at low and large signal.

I updated your model to extract the angle you need for the stability analysis.

Also the low frequency oscillation in speed and electrical power are here

• Here the simulated fault happen and cleared at subcylce (0.05 cycle) You should study your network to find the critical clearing angle of the fault

due to clearing the fault at earlier time (study the critical angle of the system) there is only swinging and hunting which can be only damped by PSS, speed controllers and Excitation system.

Finally: changing the fault clearing time affect the electrical angle oscillation.

changing the type of fault (single phase or two phase or three phase ) to ground or through resistance affect affect the stability.

Adding speed controller and excitation system will stabilize the machine and mitigate low frequency oscillation which appeared in the modification.

I wish my answer be oriented to your question.

[vincent cote]

thanks for this quick reply.

I cannot open your model because i only have matlab 2015a. Can you upload it in this version?

First of all, you are right i am actually talking about de fault at B3 who occur at 0.5 sec for 6 cycles. This fault should keep the system stable and the angle should reach about 1.07 rad (mecanical rotor angle). The real issue is that i don't want to damp my system.. is for education purpose and i want the students to see the systeme with damping and without damping. I think i was not clear enough. We can see the problem in theses pictures, the mecanical angle (rotor) diverge:

Here you can see that the angle diverge. If i zoom closer.

I had run the same model on ePPHASORsim (plugin software from opal-rt for simulink real-time simulation) and this divergence dont appear. Im scared because the simulation will run in real-time. If the simulation run for a long period, the results will be affected

Thanks for your help this is really appreciate

looking forward to your reply

Vincent