I’m afraid I don’t have any details on how these particular parameters were 
calculated, but I would not be surprised if any physically unrealistic values 
arise from equivalencing. I think there may be some negative resistances in 
some of the cases as well, which I’m pretty sure comes from equivalencing.

   Ray


On Nov 25, 2021, at 1:17 PM, Liang Chen 
<[email protected]<mailto:[email protected]>> wrote:

Dear all,

I noticed some lines with very high (1.0 and above) R/X values in certain test 
cases For example,

  1.  in case300.m, branches
     *   (9021, 9023) R/X ratio = 1.1617
     *   (9003, 9044) R/X ratio = 1.1615
     *   (9044, 9004) R/X ratio = 1.3241
  2.  in case2869pegase.m, branches
     *   (3095, 2951) R/X ratio = 1.8244
     *   (8209, 1998) R/X ratio = 1.4405
     *   (1216, 5233) R/X ratio = 1.1316
     *   (4318, 1338) R/X ratio = 1.0228
     *   (6047, 5271) R/X ratio = 1.0783
     *   (6047, 5271) R/X ratio = 1.5017

I'm wondering why this is the case and whether they are physically realistic.

I'm working on an iterative power flow algorithm and have found that high R/X 
ratio (1.0 and above) in the branches causes my iterations to diverge, so I 
want to decide between going back and revise the theory, or to simply put a 
footnote, e.g., "R/X ratio should not be too high." Numerically, capping these 
lines at R/X = 0.999 still allows for convergence, but more slowly as the 
maximum gets closer to 0.999.


Best regards,

Jeffrey Chen


Reply via email to