Dear Michael,
I try to give you some answers from the point of view of our department of a large HVAC installation company, dedicated to energy efficiency.
We have been developing our own components for three years.
“At the moment, we think there are two drivers to move to TRNSYS. Firstly, we would be able to model systems where it is required to simulate transient behaviour..”
TRNSYS has the potential to model transient behaviours, but take in mind that most of TRNSYS components are developed to be used for “quasi-steady” modeling
“Secondly, as TRNSYS has a modular structure, we feel there may be a reduction in time required to develop models. Since EES is fully equation-based, the mathematical equation-equivalent of complex and large systems can become quite extensive and tedious to develop. It seems TRNSYS would more readily enable the re-use of sub-models in various projects and therefore reduce time to develop and check models.”
You are right : reusability of components (modularity) is one of the main strength of TRNSYS !
However, consider that developing a new submodel (component) in TRNSYS demands more effort than its equivalent in an equation based solver (EES, for example, or Modelica, which could be a serious challenger of TRNSYS in your case).
“The biggest question mark around TRNSYS is this: By researching the possibilities and available components, I am now 100% convinced we will need to develop our own component models for many of our system components and control systems. In essence, I don’t see this as a problem.”
This is the typical situation of an industrial company, where we want to model systems basing on our knowledge, our data and our functional analysis.
“. However, after reading through the several manuals, working with the demo and reading the various threads on this mailing list, it seems we can expect a lot of trouble/problems when developing components. As one example, there seems to be no clear guidance on which Fortran compiler to use. Many people recommend an old compiler (Compaq Visual Fortran) which is no longer available. There seems to be no information which details how to develop new components with for example the latest Intel Visual Fortran Composer XE 2013..”
Creating new components with the latest Intel Visual Fortran Composer XE 2013 is well documented in the last version of the programmer’s guide (chapters 7.5.5 and 7.5.6). Personally, I found no problem to create my first components.
Moreover, I think that Intel Visual Fortran is the suggested development suite for TRNSYS 17. By the way, distributed version of TrdDll, which is the core of the TRNSYS engine, is compiled using Intel Visual Fortran (even though I ignore the exact version used by the developer team). Compaq Visual Fortran is an old staff today, but it is used often because, looking a little, it can be found for free (but Visual Fortran is not expensive).
“Besides that, I can see many people struggling on these threads with errors that are seemingly impossible to correct of solve.”
I find that most of problems that are posted on this list derive from an insufficient knowledge of good programming practices in general, and of Fortran in particular. Fortran is not VBA : array limit checking is not automatic, division by zero is possible and debugging is more difficult than for an interpreted language (VBA).
“Are these merely problems related to the Fortran programming itself?”
Not only Fortran but any powerful language, which should be used knowing what you do…
“ I may be wrong, but there seems to be not much technical support for people who are required to develop their own components.”
This is a fact : TRNSYS is not a commercial suite, finding a competent technical support is not easy. However, in my experience, if you post clearly a technical question in an acceptable English, explaining your problem and what you want to do, there are always nice people answering your questions.
“As you can see, we are struggling with the above issues. I would be delighted if people could share their thoughts on this and offer some advice.”
My two cents...
Best regards,
Marcello Caciolo
Responsable Outils et Méthodes d'Analyse Energétique
Cellule Efficacité Energétique et Environnementale - Pôle Projets et Réalisations
46 Boulevard de la Prairie au Duc
BP 40119 - 44201 Nantes Cedex 02
Tél.: +33 02 40 41 06 57
marcello.caciolo@cofelyaxima-gdfsuez.com
P Pensez à l'environnement avant d'imprimer ce message