If the energy (de)stored between the initial and final state cannot explain the differences, I think there is another possible cause: Type 4 can have quite large energy balance errors when the nodes are “flushed out” during one time step.
Here is my understanding:
If you have 10 nodes and a total volume of 500 L, that means each node holds 50 L of water. Using 0.1 h time steps you cannot have a flowrate through the tank higher than 500 L/h, otherwise you will replace the entire volume of a node within a time step, and the equations used by Type 4 are not valid anymore. This would explain that it works again when you reduce the timestep (with 0.001 h you would need a flowrate of 50000 L/h to “flush” a node out).
To my knowledge this problem is specific to Type 4, i.e. it does not affect the Storage tanks in the TESS libraries (Type 534 etc.) – if you have access to them that may be a more practical solution than running 0.001 h timesteps…
Michaël Kummert
Polytechnique Montréal
From: trnsys-users-bounces@cae.wisc.edu [mailto:trnsys-users-bounces@cae.wisc.edu] On Behalf Of David BRADLEY
Sent: Thursday, October 25, 2012 17:34
To: PARTENAY Vincent
Cc: trnsys-users@cae.wisc.edu
Subject: Re: [TRNSYS-users] Type energy balance
Vincent,
When you perform an energy balance on a tank, you need to take into account its change in temperature between the beginning and the end of the simulation (output 7 of your Type4). If the tank ends up hotter at the end of the simulation than it was at the beginning, it has stored some energy. If it ends up colder, then it has given up some energy. I would imagine that with 10 nodes, there is some stratification developing in the tank that is not modeled when you have a fully mixed tank (1 node)
Best,
David
On 10/25/2012 09:40, PARTENAY Vincent wrote:
Dear all,
Typ4a (Storage tank) allows to model a stratified tank with several temperature nodes. The problem I noticed is that the energy balance of both sides on the tank can vary depending on the chosen timestep :
In the attached file, a simplified example, the tank is coupled to an energy production and energy consumption model (very simple in an equation). Thermal losses of the tank is set to zero.
When I run a simulation with one node (1m height for 0.5m3) for the tank and a timestep of 0.1hr, I get the same energy on both side (see plotter “”Power&NRJ”, Esource = 2659kWh and Eload = 2666kWh), slight difference maybe explained by initial charging period to setpoint temperature.
When I run it in the same conditions but with 10 nodes (each 0.1m), these two energy are different ! While I still have 2666kWh for Eload I have 3146kWh !
When I run it again reducing the timestep from 0.1hr to 0.001 hr, still with 10 nodes, I get 2666kWh, the energy balance is ok…
So, my question is : is it necessary to check before every bigger simulation (that can include lots of tank models) the energy balance of this model for one defined timestep and boundary conditions to the model or is there some rules to predefine adapted discretisation for one given timestep ?
Kind regards,
Vincent PARTENAY
Energies Renouvelables - Département ESE
Centre Scientifique et Technique du Bâtiment
290 Route des Lucioles BP 209 – F 06904 Sophia Antipolis Cedex
Tél : +33-(0)4-93-95-67-21
Fax : +33-(0)4-93-95-64-31
Évitez d’imprimer vos emails inutilement. Respectez l’environnement
_______________________________________________
TRNSYS-users mailing list
TRNSYS-users@cae.wisc.edu
https://mailman.cae.wisc.edu/listinfo/trnsys-users
--
***************************
David BRADLEY
Principal
Thermal Energy Systems Specialists, LLC
22 North Carroll Street - suite 370
Madison, WI 53703 USA
P:+1.608.274.2577
F:+1.608.278.1475
d.bradley@tess-inc.com
http://www.tess-inc.com
http://www.trnsys.com