There isn't any conversion being done by MATPOWER. The important thing is that the units of power are in MW. The units of the cost function are in u/t, where u can be any unit of currency (e.g. US dollars) and t can be any unit of time (e.g. hours). This implies that the linear term of the cost function (and all of the prices in the results) would have the units of u/(MW * t).
So, again, the important thing when creating your cost functions f(p) is that p is in MW, while u and t can be whatever you find useful, keeping in mind that the prices in the results will be in terms of the same units. The labeling in MATPOWER's printed output assumes that u is $ and t is hours. Hope this clears up any confusion. -- Ray Zimmerman Senior Research Associate 211 Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On May 27, 2010, at 7:23 AM, Malcolm Barnacle wrote: Hi Ray, Sorry but is that definately true about the units of b being $/MWh? The reason why I am asking again is because when looking at the gencost input of case2883wp (polish network at winter peak) the numbers for this are similar in size as the numbers for my UK optimal power flow. For my UKOPF, I converted my p(pence)/KWh generation costs into p/MW using various capacity factors. I did this as I was sure that because P (active power) was in MW, b must be in p(or $)/MW aswell in order for the cost polynomial to work properly. Are you saying then that Matpower takes the p(or $)/MWh, as inputted for generation cost, and converts to p(or $)/MW before applying the cost polynomial? If this is not the case, how does the cost polynomial for matpower work? Kind Regards Malcolm Barnacle ________________________________ From: [email protected]<mailto:[email protected]> [[email protected]] On Behalf Of Ray Zimmerman [[email protected]] Sent: 25 May 2010 15:47 To: MATPOWER discussion forum Subject: Re: some coefficients On further thought, the units of the objective function are in $/hr, so that means the units of b would be $/MWh. Ray On May 25, 2010, at 10:31 AM, Ray Zimmerman wrote: 1) Units of b are $/MW (or $/MVAr). 2) P values are in MW (or MVAr). I will clarify this in the documentation for the next release. -- Ray Zimmerman Senior Research Associate 211 Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On May 25, 2010, at 9:01 AM, 王玉荣 wrote: Thank you. I agree with you both, and in fact, I changed the data based on the gencost matrix. But I have two more questions: 1) what are the units of this parameters? such like $/MW or M$/MW for b ? because I also want to know the cost of Var devices and moreover, I need to put these in my research, so I think I need to make sure these parameters are compatible. 2)I want to double check with you: when using f=a*P^2+b*P+c, the P values are norminal value or p.u.? I used norminal value. Thanks. On Mon, May 24, 2010 at 10:46 PM, Ray Zimmerman <[email protected]<mailto:[email protected]>> wrote: I would just caution you that, just because you find a case file with a particular set of cost coefficients does not mean that those cost coefficients are in any sense "typical" generator costs. In many cases, they are more likely some random numbers that someone typed in so they could run an OPF. -- Ray Zimmerman Senior Research Associate 211 Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On May 24, 2010, at 8:38 PM, m d wrote: Hello, I have just got familiar with Matpower too but as much as I know in every casedata there is a part namely: generator cost data that you can find the relevant cost coefficients. for example type: open case9 then in case9.m file at the end of the page in the generator cost data there is a matrix: mpc.gencost with 7 columns...the last three columns are the coefficients that you need. should you need more details, just simply type : 'help caseformat ' and you will find good information. hope it helps ________________________________ From: 王玉荣 <[email protected]<mailto:[email protected]>> To: [email protected]<mailto:[email protected]> Sent: Sun, May 23, 2010 2:50:53 PM Subject: some coefficients Hi All, I am doing optimal research recently, and I badly need some coefficients related to the generation cost and some VAr devices cost. Specifically, the generation cost is like this: f=a*P^2+b*P+c, where the P is the active power output. The coefficients are a,b,c. For the Var devices cost, I hope the cost is composed by f=C_fix+C_variable*Qc, where Qc is the output reative power of the Var device. And the coefficients are C_fix and C_variable. Does anybody have some reference data please? Thank you so much. Regards, Yurong -- 王玉荣 (Yurong WANG) Dept. of EECS, the University of Tennessee Knoxville, TN37996, USA School of Electrical Engineering Southeast University, Nanjing, China 210096 We Will Progress. -- 王玉荣 (Yurong WANG) Dept. of EECS, the University of Tennessee Knoxville, TN37996, USA School of Electrical Engineering Southeast University, Nanjing, China 210096 We Will Progress.
