Time-Dependent Heat Transfer/Convection Simulation of a building: Singularity failure

Please login with a confirmed email address before reporting spam

Hello everyone,

I'm currently trying to simulate the heating of a building with particular focus on its surroundings, throughout an entire summer day. For this, I am using a 24 hours time-dependent study, including the heat transfer and the laminar flow physics modules. When first approaching this project, I focused on the building heating up due to the sun's irradiation, making use of the heat transfer and the surface-to-surface radiation modules and omitting the laminar flow module. This has worked out fine and yielded usable results.

For simplification, I modeled two (outer) walls of a building, a ground floor and the air above. Note that the ground floor and the air are supposed represent the outside of the building.

Sadly I encountered severe problems when trying to simulate the convection of the air outside the building. The computation of my current model fails with the error Repeated error test failures. May have reached a singularity.. The error occurs as soon as the temperature of the building walls diverges from its initial values; however given that the building walls "must" heat up over time, I do not know what I need to change in my model to solve this error.

I wasn't able to reduce my .mph file to the required < 5 MB in order to upload it. If someone here would like to look at it anyway, I will gladly find another way to upload it somewhere else. For now, I will list the most relevant information below and include another screen shot of my model tree to give a basic idea of my physics module usage.

I would be very happy to receive some pointers to where exactly my problems stem from and what I could do to solve them. If someone here has more experience with modeling heating and convection, I finally would like to ask if my particular simulation case could also be approached with the "indirect" convection modeling techniques from Modeling Natural and Forced Convection in COMSOL Multiphysics® or through the One-Way Coupled Studies for Nonisothermal Flow.

Thank you very much and best regards

Luca

=============== Additional information: * Irradiated surface temperature: From my previous model using only the ht and rad modules, I recorded the resulting average wall and ground floor surface temperatures throughout the simulated day duration to a table. In my current model, I use this table to statically set the surface temperature for each simulated point of day. I do this over the ht > Boundary Temperature node. The range is from 12 °C to 55 °C. * Distant air temperature: In order to represent the temperature of the surrounding air, I similarly set the surface temperature of the air at the outer edges of my model through a simple diurnal cycle ranging from 12 °C to 27°C. * Mesh: I use a physics-controlled finer mesh. * Study: I use a time-dependent study with the following output times, in seconds except where specified differently: range(0,0.1,1), range(1,1,10) range(10, 30min, 23.5h)



Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.