I do not understand the overall structure of the problem you are trying to solve (and don't really want or need to). However, you claim to be running a power flow which may or may not result in violations. The issue is, how is the dispatch for that power flow determined? If it comes from a process that does *not* take into account thermal limits, then you should expect that they will at times be violated. If you want "fix" these thermal violations at the time of running the power flow, then you will need to redispatch the system somehow and you need to know what criteria you will use to determine that redispatch. Specifically, what are you attempting to preserve as you do the redispatch?
We may be able to help you implement something once we understand what you *want* the power flow to accomplish for you. -- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Jan 18, 2012, at 11:37 AM, iman wrote: > Dear Dr Zimmerman, > > > > My optimization problem determines generation capacity and location based on > the results from Matpower power flow(by calling a function). After iterations > are stopped, I looked at the slack bus branch values in Command Window and > saw that I have 122.54 from bus 1 to 2 (due to 189.98 MW generations in slack > bus) which is a violation for the branch. Now I also have the results from my > optimization giving bus 19 and 8 as the optimal places. > > To my dismay if I add two DGs (got from the last step ) into the locations > and I run PF again I don’t see this violation! > > Example: my results says optimal capacity for bus 19 and 8 are 32 and 47 MW,I > changed the mpc.gen and got -40.97 from the bus 1 to 2 and -53.33 for the > slack bus generation (No Violation anymore!!) > > > > Do you thing that there might be a problem with my code? Or does that make > sense? > > > > On Wed, Jan 18, 2012 at 3:27 PM, Ray Zimmerman <[email protected]> wrote: > Iman, > > If I understand correctly, your optimization program gives you a dispatch > that is optimal in some sense but results in violated thermal limits and you > want to redispatch the generation to respect the thermal limits. Correct? > > If so, what criteria do you propose to use to determine this redispatch? > > -- > Ray Zimmerman > Senior Research Associate > 419A Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > > On Jan 18, 2012, at 9:13 AM, iman wrote: > >> Dear Dr Zimmerman, Shiyang, Li and Arash >> >> >> >> Thank you for your reply to my problem. Dr Zimmerman has said that I need to >> use OPF instead of PF for branch thermal constraints. The problem is that I >> am using another optimization method connected to Matpower and this >> optimization program only needs to use V,P,Q and delta values from the power >> flow calculations. In my optimization program, I am trying to minimize P >> loss and other things so if I use OPF, my final results would be affected >> and they are not purely from my optimization method. >> >> I also tried to code the thermal limit for PF myself but I haven’t been able >> to. >> >> Would you please help me in coding the thermal constraints? >> >> >> >> Regards >> >> Iman >> >> >> >> 2012/1/13 新浪VIP <[email protected]> >> Hi Iman， >> 1, Even your object is to study the system near static limit, keeping all >> generation of PV buses is usually not reasonable, because this scenario may >> never occur. For this sake, when we calculate load margin for example, we >> normally increase P generation with loads. >> >> 2, Power is always conserved on static state. Without PV generation >> increasing >> , the slack bus has to compensate all the deficit, 236.55 MW here. >> >> 3, Like what Dr Ray said, if you want to involve thermal limits of branches, >> you may either use OPF solver, or code you own function to adjust PV >> generations. >> >> Shiyang Li >> >> 发自我的 iPhone >> >> 在 2012-1-12，21:44，iman <[email protected]> 写道： >> >>> Dear Shiyang, Li >>> >>> Thanks for taking time answering. >>> >>> I didn’t increase any generations in the PV buses because my objective for >>> increasing the load is pushing the system to the instability margin (like >>> voltage drop in buses). >>> >>> I agree with you that network loss doesn’t seem a real number as well as >>> other numbers. >>> >>> In the case30.m (base case) after running pf I have: >>> >>> Total generation capacity =335 MW >>> >>> On-line capacity =335 MW >>> >>> Continued Gens (actual generation)=191.6 MW >>> >>> Load = 189.2 MW >>> >>> Losses=2.44 MW >>> >>> >>> >>> Branch 1-2 =10.89 MW >>> >>> Branch 1-3=15.08 MW >>> >>> ****Now I multiply Pd (demanded power)by 2 to make the whole consumption >>> twice, I have*** >>> >>> Total generation capacity =815 MW >>> >>> On-line capacity =815 MW >>> >>> Continued Gens (actual generation)=402.2 MW >>> >>> Load = 378 MW >>> >>> Losses=23.82 MW >>> >>> >>> >>> Branch 1-2 =154.81 MW >>> >>> Branch 1-3=81.08 MW >>> >>> >>> >>> >>> >>> Having said the above in brief: >>> >>> 1-In Bus 1 (Slack) we have 236.55 MW injection power .Where does this added >>> amount come from? >>> >>> 2-What if I want to fix branch 1-2 limit to 100 MW (or 100 MVA)? What >>> should I do? >>> >>> >>> Thank you for your attention and time >>> >>> Wish you the best. >>> >>> Iman >>> >>> >>> >>> On Wed, Jan 11, 2012 at 7:45 PM, 李诗旸 <[email protected]> wrote: >>> Dear Iman, >>> >>> With the load increasing, have you increase the generation from PV buses? >>> I don’t think the network loss could usually increase to nearly 10 times >>> even at static limit. You may double the Pg on PV buses with the load. >>> Then, if you still want to impose the branch thermal limits, you have to >>> code the function to adjust the Pg for PV buses based on participant >>> factors (branch flow sensitivity w.r.t. generation) or others. >>> >>> >>> >>> Shiyang, Li >>> >>> >>> >>> >>> >>> 发件人: [email protected] >>> [mailto:[email protected]] 代表 iman >>> 发送时间: 2012年1月12日 0:57 >>> 收件人: [email protected] >>> 主题: limits on the branches >>> >>> >>> >>> Dear all, >>> >>> >>> >>> I want to use power flow (PF) for the load increase studies. I picked 30 >>> bus system with Pg=23.54 MW for the slack bus. Running the power flow with >>> the base loads ,I get 25.97 MW which is quite close to 23.54. >>> >>> Now I want to increase all of the loads 2 times. Running pf I get 236.55 MW >>> for the slack bus . >>> >>> I think in practice thermal line limits doesn’t allow this amount of power >>> dragging from slack bus. >>> >>> I want to make my problem practical by imposing thermal limits on the >>> branches .How can I do that? >>> >>> >>> >>> Thank you in advance for your patience >>> >>> I look forward to hear from you guys >>> >>> Iman >>> >>> >>> >>> >>> -- >>> Best regards >>> Iman >>> >> >> >> >> -- >> Best regards >> Iman >> > > > > > -- > Best regards > Iman >
