OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

emix-comment message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Clarification request concerning the use of the 'PowerRampSegmentType' in EMIX 1.0



Specification Reference:

EMIX 1.0     13.6.1 Power Ramp Segments


Background:

The SPC201P FSGIM standard extensively references EMIX.  During a FSGIM review cycle, a comment was received requesting that the FSGIM data structure defined for describing ramping demand (DPieceWiseLinear) be replaced with the analogous EMIX ramping construct.  This construction appears to be PowerRampSegmentType as shown below immediately followed by the UML description.



The UML model reads:



where:

rate - the power component of a ratio specifying the maximum rate of change of power for this segment.
duration - the time period component of a ratio specifying the maximum rate of change of power for this segment.
beginRamp - Quantity of power at the start of a ramp segment.
endRamp - Quantity of power at the end of a ramp segment.
integralOnly - If true, offer is all or none; no partial acceptance of this segment.



Items Needing Clarification:


There are several questions concerning this construction:

1) The definition of the 'rate' attribute differs (or appears to differ) between the text and UML versions of the model.  The text seems to indicate that the 'rate' describes the units of the line segment; the UML model indicates it defines the slope of the line segment.  The actual data type construct seems to define both in 'powerItem' and 'quality' attributes respectively.  

The question is,  "What is 'rate' used for?"  Is it supposed to represent (End Ramp Power - Begin Ramp Power)or the units for those power quantities?  If it's the former, then please change the Note in the documentary specification to say that (and possibly change the name of the element to better indicate it).  If it's the latter, then please change the name of the Element to "Unit" and its type in the schema to PowerItemType.

2) The line segment slope is given by either the algorithm [(endRamp - beginRamp)/duration] or by the rate.quantity attribute directly.  It seems redundant to have multiple attributes defining the same value.

3) beginRamp and endRamp are typed as 'integer'.  I would expect that these attributes would be at least typed as 'float'.  Better yet, I would expect they would be typed to handle real and apparent power values.

4) The PowerItem attribute is defined for each line segment of the ramping curve thereby requiring each line segment to describe an 'item description',  'item units, 'siScaleCode', hertz, voltage and AC.  It seems this meta data should be promoted up a level so it is defined one time for the entire ramping curve.  Furthermore, there are no multiplicities defined for this meta data.  Is this implicitly noting these are required attributes (i.e. [1..1]) or does EMIX not define multiplicities across the specification?



If you can provide insight to these questions, I would like to use this construct in the SPC201P model.


Regards,

Jerry Martocci




[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]