Using GA for any type of optimization using power flow is asking for troubles and especially so with distribution systems. At first it seems straight forward, easy, and simple if you can live with a small inaccuracies of the optimal point. However, if GA is using power flow to search for the optimal point, it will run into infeasible regions or simply regions where power flow fails numerically and sometimes you might not even know it. If you can determine in advance that the search domain is feasible and numerically stable, GA is a great approach. If you can determine the feasibility regions beforehand, I would be very interested how and will share my experience with you. This is a serious problem with equivalent systems and I would expect distribution systems as well. If you can, use some kind of decoupled power flow (it should be OK for planning purposes) and subsequently check for feasibility. Of course, there are not too many decoupled power flows for distribution systems around but you could again check Ray's PhD.
Ray Zimmerman wrote: > MATPOWER does not include the backward/forward power flow method for > radial systems. > > -- > Ray Zimmerman > Senior Research Associate > 428-B Phillips Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > On Jul 16, 2007, at 12:25 PM, ali jahanbani ardakani wrote: > >> Hi All. >> I am working on planning of Distribution networks using GA for my MSc >> Thesis. >> >> >> In performing Load flow I encountered a problem when using matpower. >> In Radial networks the newoton raphson method converges and I couldn't >> obtain Voltages, I have asked some people, they say I must use forward >> backward loadflow, >> Does someone has it's code in MATLAB M-File?????? >> I am really in need. >> please send me it if you have it, >> Thanks beforehand >> thank you all >> >> goodbye >
