From GridLAB-D Wiki
Jump to: navigation, search

Ftuffner 01:45, 28 February 2013 (UTC)

  • Under output format, I'm not sure what a "chile object" is
  • You should clarify what the various "ZCA" and "Zph1ph1" references are for the specification
  • I'm hoping it is just your example, but 2 digits past the decimal for the output is a little lax. IEEE test systems usually have at least 4.
  • The group_id field seems to indicate that it is required. If not specified, wouldn't we want it to dump all impedances of all line objects?
  • You mention hooks are there for transformers and regulators further down the road. I think you should explicitly state how these will be handled, even if we aren't coding them up. Namely, what is this impedance value output? Is this source-side, load-side, or both-side referenced?
  • How do you plan to handle "asymmetric" impedance values? Namely, for something like a Delta-GWye transformer and SPCT transformer, its impedance isn't a turns-ratio scaled version when transitioning between source-side and load-side values.
  • You may want to define an optional "voltage ratio" field for the future transformer/regulator implementation. Even if it isn't coded, that shows the information was considered.
  • How are you outputting single-phased or two-phased objects? I assume with just zero values in the appropriate matrix elements, but it may be useful to define this (or has some form of "empty" designation).
  • I'm not sure I agree with the ".csv output" suggestion. As written, the first element will be merged with your fields, so that won't be very useful to extract. If you're going to promote a CSV structure, I'd almost suggest doing the whole matrix as a single row of CSV outputs. e.g.,
 name,type,from,to,z11,z12,z13,z21,z22,z23,z31,z32,z33
 test,overhead_line,node_a,node_b,1.1001+j2.50,...
  • It may be useful for a length field in here as well, so base impedance values can be extracted if desired
  • Does any information on the "source" of this impedance information need to be handled, or is that beyond the details of this spec?