Cedric,
I am looking into this. It is important to keep in mind that the finite difference algorithm used by the slab preprocessor is not unconditionally stable. I wouldn’t read too much into the meaning of the 580°C swing, that’s just the nature of numerical instability. That said, I suspect that the problem you are seeing is an artifact of the automatic domain discretization. It may be that one of the differential elements is too thin for a stable solution. Changing material properties can also cause problems for stability. The insulation values you are testing seem rather high – keep in mind that these should be in SI units (m2-K/W) and usually won’t be higher than 5 m2-K/W.
Neal
From: EnergyPlus_Support@xxxxxxxxxxxxxxx [mailto:EnergyPlus_Support@xxxxxxxxxxxxxxx] On Behalf Of cedricdlb
Sent: Friday, April 22, 2011 12:52 PM
To: EnergyPlus_Support@xxxxxxxxxxxxxxx
Subject: [EnergyPlus_Support] GHT Slab severe errors and high sensitivity to inputs
The Ground Heat Transfer to Slab calculation gives Severe (Fatal) errors and exhibits extreme and unpredictable sensitivity to inputs. I would like to know how to programmaticly adjust my inputs to prevent these Severe errors. I'm using Energy Plus version 6 and the Auxiliary Slab program. The results are the same whether I use the pre-processor invoked by EnergyPlus or the slab stand-alone program.
Below my question is an example idf file for stand-alone input into the slab utility. I used this file with the San Francisco weather file (USA_CA_San.Francisco.Intl.AP.724940_TMY3.epw). Given this idf, even slight variations in the value of the EquivalentSlab's APRatio (Area Perimeter Ratio) can lead to extreme temperature swings, even values below absolute zero (-273 C) for the calculated ground temperature. For instance:
Program calculated
Status temperature APR
------- ----------- ---
Severe -352.01 C 2.50123157894737 (value in attached idf)
Severe 230.77 C 2.501 (minute change produces wild 580 degree C swing)
Success ~19.46 C 2.50 (temp is average of TAverage, TPerimeter, & TCore values)
Success ~19.47 C 2.51 (it wasn't that 2.501 was too large, since 2.51 works fine)
Success ~19.46 C 2.495 (digits of precision do not appear to be the factor)
If I hold the APR at its original value (2.50123157894737), and vary the insulation, I also get wide and non-linear variations:
Status Temp Insulation R-value for 1 inch thick under-slab
------ --------- ----------------------------------------------
Severe -284.58 C 20.0
Severe -327.62 C 22.0
Severe -364.49 C 25.0 (local minimum, lower temp than more or less insulation)
Severe -331.38 C 30.0
Is there some factor in there that I need to keep within a certain value or precision range, to ensure successful runs?
Thank you,
Cedric de La Beaujardiere
Here are my idf inputs to the slab stand-alone program:
Materials,
2, !- NMAT: Number of materials
0.158, !- ALBEDO: Surface Albedo: No Snow
0.379, !- ALBEDO: Surface Albedo: Snow
0.9, !- EPSLW: Surface Emissivity: No Snow
0.9, !- EPSLW: Surface Emissivity: Snow
0.75, !- Z0: Surface Roughness: No Snow
0.03, !- Z0: Surface Roughness: Snow
6.13, !- HIN: Indoor HConv: Downward Flow {W/m2-K}
9.26; !- HIN: Indoor HConv: Upward {W/m2-K}
MatlProps,
2243.0, !- RHO: Slab Material density {kg/m3}
1200.0, !- RHO: Soil Density {kg/m3}
837.0, !- CP: Slab CP {J/kg-K}
1200.0, !- CP: Soil CP {J/kg-K}
1.8, !- TCON: Slab k {W/m-K}
2.0775; !- TCON: Soil k {W/m-K}
BoundConds,
TRUE, !- EVTR: Is surface evapotranspiration modeled
TRUE, !- FIXBC: is the lower boundary at a fixed temperature
10, !- TDEEPin {C}
False, !- USRHflag: Is the ground surface h specified by the user?
; !- USERH: User specified ground surface heat transfer coeff {W/m2-K}
BldgProps,
10, !- IYRS: Number of years to iterate
0, !- Shape: Slab shape
2.4384, !- HBLDG: Building height {m}
25.555555555556, !- TIN1: January Indoor Average temperature set point {C}
25.555555555556, !- TIN2: February Indoor Average temperature set point {C}
25.555555555556, !- TIN3: March Indoor Average temperature set point {C}
16.9, !- TIN4: April Indoor Average temperature set point {C}
21.111111111111, !- TIN5: May Indoor Average temperature set point {C}
21.111111111111, !- TIN6: June Indoor Average temperature set point {C}
21.111111111111, !- TIN7: July Indoor Average temperature set point {C}
21.111111111111, !- TIN8: August Indoor Average temperature set point {C}
21.111111111111, !- TIN9: September Indoor Average temperature set point {C}
25.555555555556, !- TIN10: October Indoor Average temperature set point {C}
25.555555555556, !- TIN11: NovemberIndoor Average temperature set point {C}
25.555555555556, !- TIN12: December Indoor Average temperature set point {C}
0.0, !- TINAmp: Daily Indoor sine wave variation amplitude {deltaC}
0.1; !- ConvTol: Convergence Tolerance
Insulation,
0.0, !- RINS: R value of under slab insulation {m2-K/W}
0.0, !- DINS: Width of strip of under slab insulation {m}
0.0, !- RVINS: R value of vertical insulation {m2-K/W}
0.0, !- ZVINS: Depth of vertical insulation {m}
0; !- IVINS: Flag: Is there vertical insulation
EquivalentSlab,
2.50123157894737, !- APRatio: The area to perimeter ratio for this slab {m}
0.101, !- SLABDEPTH: Thickness of slab on grade {m}
15.0, !- CLEARANCE: Distance from edge of slab to domain edge {m}
15.0; !- ZCLEARANCE: Distance from bottom of slab to domain bottom {m}
__._,_.___
Primary EnergyPlus support is found at:
http://energyplus.helpserve.com or send a message to energyplus-support@xxxxxxxx
The primary EnergyPlus web site is found at:
http://www.energyplus.gov
The group web site is:
http://groups..yahoo.com/group/EnergyPlus_Support/
Attachments are currently allowed but be mindful that not everyone has a high speed connection. Limit attachments to small files.
EnergyPlus Documentation is searchable. Open EPlusMainMenu.pdf under the Documentation link and press the "search" button.
Change settings via the Web (Yahoo! ID required)
Change settings via email: Switch delivery to Daily Digest | Switch to Fully Featured
Visit Your Group | Yahoo! Groups Terms of Use | Unsubscribe
__,_._,___
=======
Email scanned by PC Tools - No viruses or spyware found.
(Email Guard: 7.0.0.21, Virus/Spyware Database: 6.17360)
http://www.pctools.com
=======