Skip to main content


Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » Papyrus » Q about structure of ISO 80000 and QUDV in SysML Papyrus and how to apply it.(SysML, ISO 80000, QUDV, Papyrus)
Q about structure of ISO 80000 and QUDV in SysML Papyrus and how to apply it. [message #1843806] Thu, 19 August 2021 07:31 Go to next message
Marnix Verkammen is currently offline Marnix VerkammenFriend
Messages: 4
Registered: August 2021
Junior Member
Dear all,

I relatively limited experience with SysML, but I am part of a team creating some form of mapping between a SysML software and another MBSE tool.
This mapping needs to be able to interpret the ISO 80000 package, but for this, I first need to know how this package works in Papyrus, and then also better understand its structure.
I am using the ISO80000 library given in Commnent 6 of this bug-thread: Comment 6 of this Eclipse bug discussion: https://bugs.eclipse.org/bugs/show_bug.cgi?id=478131#c6


I am unsure as to how the new units or QuantityKinds (QKs) should be defined within Papyrus.
Let's assume that I want to create the unit Newton from scratch (not using the one in the ISO 80000 package), stating that it is a unit derived from kg, m and s (km/s^2).
Do I then need to create and InstanceSpecification with as name Newton and as classifier DerivedUnit (taken from the QUDV modelLibrary)?

I have tried creating a normal SysML unit, than change its classifier to derivedUnit, but then the unit transforms itself into an InstanceSpecification. Hence why I think one should create an instanceSpecification.

My question then is: how does one assign the factor, unit and (primary)quantitykind, as shown in Figure E.9 of OMG' s System Modelling Language v1.5 document? I'll place the figure below.
I see that these relationships exist in the QUDV modelLibrary in Papyrus, however I am unsure how to drag them (?) under the newton Derived Unit.

Would someone be so kind as to give more information, or a resource in which this can be found? I have searched quite extensively on the internet, but have yet to find a manual, forum thread, etc. about using the QUDV library in Papyrus.

----------------------------------------------------

A second question I have regards the structure of the pre-existing Units, Valuetypes and QuantityKinds. Are they linked properly in the Papyrus ISO 80000 package? By that I mean for instance: Are the unit and quantityKinds of a ValueType properly defined? I see that most (if not all, since I have not checked all of them of course) Valuetypes have empty Units and QKs. An example of this is <<ValuetType>>mass(zeptogram), see image below.

In the same line of thought: Are prefixed units (or derived units for that matter) actually defined as described in the OMG SysML document? As far as I can see prefixedUnits such as zeptogram have two slots: one for the name (in an instanceValue) and one for the prefix. The latter seems to be also an instanceValue with "zepto" written in it. Hence, I am not sure if it is in any way connected to the prefix "zepto" defined in ISO 80000-1

---------------------------------------------------

I look forward to your help or advice.
Kind regards,
Marnix.

PS: I originally posted this topic in the "Papyrus for Real Time" forum, but I was informed that this forum is best for this specific topic. If that would not be the case, please let me know.
Re: Q about structure of ISO 80000 and QUDV in SysML Papyrus and how to apply it. [message #1843813 is a reply to message #1843806] Thu, 19 August 2021 12:44 Go to previous messageGo to next message
Patrick Tessier is currently offline Patrick TessierFriend
Messages: 341
Registered: July 2009
Location: Paris Saclay, France
Senior Member
Hello,
I have tried to create the Derived Unit Newton. Wow It is very long to do. you have to create an instance spec , choose the classifier Derived Unit from the QUDV library, and then create slots that corresponds to each interesting feature.
In papyrus we have no tool create automatically slots from a referenced classifier inside the instanceSpec.
So it is possible but it will take a very long time.
  • Attachment: Newton.PNG
    (Size: 37.41KB, Downloaded 108 times)

[Updated on: Thu, 19 August 2021 12:45]

Report message to a moderator

Re: Q about structure of ISO 80000 and QUDV in SysML Papyrus and how to apply it. [message #1843823 is a reply to message #1843813] Thu, 19 August 2021 15:53 Go to previous messageGo to next message
Marnix Verkammen is currently offline Marnix VerkammenFriend
Messages: 4
Registered: August 2021
Junior Member
Hey Patrick,

Thanks for trying to create a newton Derived Unit! Indeed, it seems to me as well that this is a tedious manual process. Thanks for confirming it!

However, in your Newton, I see you have only added a Slot for the symbol, with its Value. If you look at the Figure E.9 which I shared in my first post, I am wondering how you would go on to create the "factor" dependency linking Newton with one of its UnitFactors? And, especially how you envision creating the "unit" dependency linking this unitFactor with its "SimpleUnit". Something similar could be said about the "quantityKindFactor" and the "SimpleQuantityKind", and their dependencies.

I see a possibility of creating a slot with as Defining Feature the "UnitFactor" property of the DerivedUnit, given in the QUDV modelLibrary. Then, a value of "1/1" could be given (if we are looking at the factor of mass for instance). But: Then you have the issue of how one should link this "Unitfactor" slot with a "SimpleUnit" called mass... As far as I can see, one cannot create a slot within a slot. And, I am not sure how to use the "A_unitFactor_unit" and the other dependencies defined within the QUDV modelLibrary.
Would you have any suggestions for that?



My second question also still holds, by the way:
Quote:
second question I have regards the structure of the pre-existing Units, Valuetypes and QuantityKinds. Are they linked properly in the Papyrus ISO 80000 package? By that I mean for instance: Are the unit and quantityKinds of a ValueType properly defined? I see that most (if not all, since I have not checked all of them of course) Valuetypes have empty Units and QKs. An example of this is <<ValuetType>>mass(zeptogram), see image below.

In the same line of thought: Are prefixed units (or derived units for that matter) actually defined as described in the OMG SysML document? As far as I can see prefixedUnits such as zeptogram have two slots: one for the name (in an instanceValue) and one for the prefix. The latter seems to be also an instanceValue with "zepto" written in it. Hence, I am not sure if it is in any way connected to the prefix "zepto" defined in ISO 80000-1


But, to me, it seems like everyuthing is very loosly defined, isn't it? Slots seem to be the main way that was used to define the Units, QKs and ValueTypes, without clear interconnections.

Regards,
Marnix
Re: Q about structure of ISO 80000 and QUDV in SysML Papyrus and how to apply it. [message #1843824 is a reply to message #1843823] Thu, 19 August 2021 16:32 Go to previous message
Carsten Pitz is currently offline Carsten PitzFriend
Messages: 479
Registered: May 2015
Location: Germany
Senior Member
Hi Marnix,
depending on the number of derived units to add it might pay out to code a little generator that generates an Papyrus *.uml file directly. But I do not advise you to do so if the number of units to add is below 100.
/Carsten
Previous Topic:Help: how to generate XML file corresponding to UMl model
Next Topic:Cannot install Papyrus-Update-5.2.0I2.zip on 2021-06 - missing XWT
Goto Forum:
  


Current Time: Tue Apr 23 16:15:53 GMT 2024

Powered by FUDForum. Page generated in 0.03261 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software

Back to the top