Hello, I am trying to estimate a frequency of power system at the moment, but I have got some problems with methods I have used. Last week I found out, the smart DFT exists. I hope, it can be a appropriate way for me. But on the net there is not much information about this, I mean, very interesting part of DFT. Can somebody explain me the algorithm of smart DFT? Thank you very much. Mira, Czech Republic

# Smart DFT

Started by ●March 11, 2013

Reply by ●March 11, 20132013-03-11

On 3/11/13 8:26 AM, Mimar wrote:> Hello, > > I am trying to estimate a frequency of power system at the moment, but I > have got some problems with methods I have used. > > Last week I found out, the smart DFT exists. I hope, it can be a > appropriate way for me. But on the net there is not much information about > this, I mean, very interesting part of DFT. > > Can somebody explain me the algorithm of smart DFT? >reduced calories? Apple or Android operating system? not George W Bush? gee, i dunno. -- r b-j rbj@audioimagination.com "Imagination is more important than knowledge."

Reply by ●March 11, 20132013-03-11

> On 3/11/13 8:26 AM, Mimar wrote: > > Hello, > > I am trying to estimate a frequency of power system at the moment, but I > have got some problems with methods I have used. > > Last week I found out, the smart DFT exists. I hope, it can be a > appropriate way for me. But on the net there is not much information about > this, I mean, very interesting part of DFT. > > Can somebody explain me the algorithm of smart DFT? >Apparently, it's a term from power systems. There actually are some references to a 'smart' DFT (or SDFT). It seems it has to do with estimating the fundamental line frequency after faults: http://bu.umbb.dz/ressources-electroniq/Bibumbb/publication/article%20pdf/Hamid%20Bentarzi%20,%20Abderrahmane%20Ouadi%20et%20J.C.%20Maun.pdf The 'smart' DFT is reference [15] cited in the above. There are other papers: http://libra.msra.cn/Publication/1531353/a-precise-calculation-of-power-system-frequency and http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=1489295&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D1489295 If you search for 'SDFT' and 'power', you can find a few more hits. Sorry if I can't tell you much about the algorithm, though � it's a recursive one, and seems to be specific to estimating the line frequency quickly after a disturbance. Kevin

Reply by ●March 12, 20132013-03-12

>> On 3/11/13 8:26 AM, Mimar wrote: >> >> Hello, >> >> I am trying to estimate a frequency of power system at the moment, butI>> have got some problems with methods I have used. >> >> Last week I found out, the smart DFT exists. I hope, it can be a >> appropriate way for me. But on the net there is not much informationabou=>t >> this, I mean, very interesting part of DFT. >> >> Can somebody explain me the algorithm of smart DFT? >> > >Apparently, it's a term from power systems. There actually are some >references to a 'smart' DFT (or SDFT). It seems it has to do with >estimating the fundamental line frequency after faults: > >http://bu.umbb.dz/ressources-electroniq/Bibumbb/publication/article%20pdf/H= >amid%20Bentarzi%20,%20Abderrahmane%20Ouadi%20et%20J.C.%20Maun.pdf > >The 'smart' DFT is reference [15] cited in the above. There are other >papers: > >http://libra.msra.cn/Publication/1531353/a-precise-calculation-of-power-sys= >tem-frequency > >and > >http://ieeexplore.ieee.org/xpl/login.jsp?tp=3D&arnumber=3D1489295&url=3Dhtt= >p%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D1489295 > >If you search for 'SDFT' and 'power', you can find a few more hits. > >Sorry if I can't tell you much about the algorithm, though =96 it's a >recursive one, and seems to be specific to estimating the line >frequency quickly after a disturbance. > >Kevin >Hello, I decided to use smart DFT, because other ways of frequency estimation were not working as well as I wanted. Last week I was testing sliding DFT with Jacobsen estimator, but I had got some problems with frequency changes. If the signal, I focused on, had got constant frequency, this algorithm provided excellent reesults. But if I started to change frequency with linear progress, it stopped working well. Some swings and bad values have appeared there. So I will test SDFT next days. Thank very much Kevin, you sent me some links. P.S.: I am sorry, my English is not good.