So it seems you could model it two different ways in MATPOWER. 1) Exclude bus GSP and the OLTC from the model and let Tx be the slack bus with a dummy generator and VMIN = VMAX = 1.078. 2) Include GSP and the OLTC, with a dummy generator at GSP (the slack bus), with VMIN = VMAX = 1.0. In this case, you would have to iteratively run the OPF, then update the tap setting until the voltage at Tx is close enough to your target. I suppose you could use VMIN = VMAX = 1.078 at Tx and then adjust the tap ratio until you get a feasible solution. You may need to leave a small epsilon difference between VMIN and VMAX at GSP or Tx in order to get feasibility.
I expect the results for the rest of the system to be (at least nearly) identical in the two cases. -- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Feb 10, 2012, at 8:32 PM, Silvio Miceli wrote: > > > The information of the network is as follows: > > The one-line diagram of a typical rural section of the Irish 38-kV > distribution network was shown in above Figure. The feeders are supplied by > one 31.5-MVA 110/38-kV transformer (capable of handling reverse power flows). > The voltage at the grid supply point is assumed to be nominal. In the > original configuration (no DG), the on-load tap changer at the substation has > a target voltage of 1.078 pu (41 kV) at the busbar, well within the +-10% > nominal voltage limits of Irish practice. > > Best Wishes > > Silvio Miceli > > > On Fri, Feb 10, 2012 at 23:07, Ray Zimmerman <[email protected]> wrote: > You haven't said which bus is your slack bus. Can I assume that it would be > the one labeled GSP? I don't see a slack generator at that bus. Is the OLTC > the *only* voltage control you have in the network? Is the voltage at GSP > fixed? > > -- > Ray Zimmerman > Senior Research Associate > 419A Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > > On Feb 10, 2012, at 2:51 PM, Silvio Miceli wrote: > >> I want to have an OLTC at slack bus only in order to control centrally the >> network voltage (active network) as below figure. How can I compare the >> results with and without OLTC? with changing tap ratio or with changing >> voltage setpoints? >> >> <image.png> >> >> Best Wishes >> >> Silvio Miceli >> >> >> On Fri, Feb 10, 2012 at 20:43, Ray Zimmerman <[email protected]> wrote: >> In order to understand clearly what you are trying to compare, I would need >> to see the network topology. >> >> But, it both cases include the OLTC in the topology and in one case you are >> modifying the tap ratio to control voltage and in the other you are simply >> modifying the generator voltage setpoints, then the two solutions will not >> be equivalent. >> >> -- >> Ray Zimmerman >> Senior Research Associate >> 419A Warren Hall, Cornell University, Ithaca, NY 14853 >> phone: (607) 255-9645 >> >> >> >> >> On Feb 10, 2012, at 10:46 AM, Silvio Miceli wrote: >> >>> Dear Ray, >>> >>> As far as I know, taking into account the voltage at slack bus as >>> optimization variable is equal to have an OLTC. So, how can I compare the >>> results with and without voltage control at slack? Can it be done either by >>> changing the tap ratio or voltage limits? >>> Best Wishes >>> >>> Silvio Miceli >>> >>> >>> On Thu, Feb 9, 2012 at 18:57, Silvio Miceli <[email protected]> wrote: >>> Dear Ray, >>> >>> 1. Can I say MATPOWER's OPF considers the power factor angle of generators >>> as optimization variable? If not, how can I consider as optimization >>> variable? >>> >>> 2. Also, according to one of your replies to a post with regards to >>> considering the slack bus voltage as optimization variable, why you want to >>> implement OLTC in MATPOWER? In my idea, considering the slack bus as >>> optimization variable is equal to have an OLTC and consequently considering >>> the secondary voltage as optimization variable. Because usually the OLTC is >>> used in order to control the voltage of slack bus and in MATPOWER is >>> already considered as optimization variable. If I am not right, please let >>> me know? >>> >>> Best Wishes >>> >>> Silvio >>> >>> >>> >>> On Wed, Feb 8, 2012 at 17:47, Ray Zimmerman <[email protected]> wrote: >>> On Feb 8, 2012, at 10:01 AM, Silvio Miceli wrote: >>> >>>> 1. What kind of generator has been taken into account in MATPOWER in >>>> Section 5.4.3 of MANUAL in order to consider the capability curve? >>> >>> It is simply intended to be a piecewise linear approximation to the kind of >>> capability curve exhibited by many types of conventional generators, such >>> as this one from Figure 2 in [1]. >>> >>>> 2. I want to minimize losses instead of maximizing social welfare >>>> considering offers and bids. How can I do it in MATPOWER? >>> >>> The answer to this one is readily available in the list archives ... e.g. >>> http://www.mail-archive.com/[email protected]/msg00817.html >>> >>>> 3. How can I maximize profit for generators in MATPOWER instead of >>>> maximizing Social welfare? >>>> Also, by which formula I can obtain profits (for generators), revenue and >>>> cost in MATPOWER? Could you please address it? >>> >>> I'm not aware of a method to maximize profits, since that would involve an >>> objective that is a function of price, a very unconventional type of >>> optimization problem. You can compute revenue directly as the product of >>> quantity and price, and the cost is available in the dispatch matrix >>> returned by runmarket. See help idx_disp for a description of each column >>> of the dispatch matrix. >>> >>> - Ray >>> >>> [1] >>> http://www.hydroworld.com/index/display/article-display/353952/articles/hydro-review/volume-28/issue-2/feature-articles/system-protection/coordinating-generator-protection-and-controls-an-overview.html >>> >>> -- >>> Ray Zimmerman >>> Senior Research Associate >>> 419A Warren Hall, Cornell University, Ithaca, NY 14853 >>> phone: (607) 255-9645 >>> >>> >>> >>> >> >> > >
