Thanks Shri for your responses. I would just add the following as a 
clarification to the second half of question 1. Unfortunately, the term "slack 
bus" is often used to refer to 2 separate concepts. One is that of a bus whose 
angle is used as a voltage angle reference. Mathematically, you should have one 
of these in every island in your network. For a fully connected network, you 
should have exactly one. The second concept is the idea of "slack" and this 
corresponds to a generator (or set of generators) whose real power is adjusted 
to match the difference between the total load+losses and the total generation 
of the other units.

So, voltage angle reference corresponds to a bus and slack corresponds to a 
generator or group of generators. Slack is only relevant for power flow, not 
optimal power flow, while voltage angle reference is needed for OPF too.

Now, in MATPOWER, setting a bus's BUS_TYPE to REF specifies that it is to be 
used as the voltage reference, i.e. that the voltage angle is to be fixed at 
the value in the input bus matrix (which is often set to zero, but not 
necessarily). It also specifies that, for power flow computations, all 
generators at the bus are to be used as slack generators. MATPOWER does not yet 
implement explicitly the idea of general distributed slack.

Hope this helps,

-- 
Ray Zimmerman
Senior Research Associate
419A Warren Hall, Cornell University, Ithaca, NY 14853
phone: (607) 255-9645




On May 8, 2013, at 7:58 AM, Shri <[email protected]> wrote:

> 
> On May 8, 2013, at 6:47 AM, ahmad rezaee wrote:
> 
>> Dear Dr Zimmerman
>> 
>> 1. A power system may have more than one slack bus?
> Yes.
>> If it may have, just the phase angle of one of them can be zero. yes?
> No, all the slack buses can have non-zero phase angles.
>> 2. In matpower, at mpc.gen at the row corresponding to "slack bus" if we 
>> change value of coloumn 2 (Pg), the powerflow results will change?
> No, Pg and Qg for the slack bus are calculated during the power flow.
> 
> Shri
>> 
>> 
>> Kind Regards
>> 
>> Ahmad
> 

Reply via email to