[Bldg-sim] Optimizing a DOAS

[Graham Carter & Megan Lyall]

With Excel you can enable circular references to enable the calculations to converge.  If you have 8760 rows then you can also run a simple macro to reseed a column with a calculated result so you in effect get convergence.

One challenge with these types of systems is what is optimal under one set of ambient and internal conditions may not be as optimal in another set of conditions.  So coming up with one set of controls will not optimise for all conditions.  None of the software that I have used over the years would allow you to properly have different controls for different ambient conditions.  So testing one set of controls and looking at annual results you would lose what is really happening under different conditions unless you look at hourly results but then your ability to adjust controls for different conditions is limited unless you go to something more open and flexible.

At the end of the day being able to model different controls would be useful to understand what the sensitivity and thus opportunity or risk is.  If the differences are small then there is a strong argument to keep it simple and find that balance of simplicity and efficiency so that long term efficiency can be sustained rather than having something no one knows how to fix or operate ...

Regards,
Graham

On 22/03/2013, at 11:32 PM, Dennis Knight wrote:

> Eric,
> I have been looking at similar situations in our warm and humid climate for about three years trying to optimize both DX and water cooled heat pump DOAS with various combinations of heat recovery (sensible and total) and have not found anything that can do a very good job of analyzing these units under all oa conditions and varying return/exhaust air conditions. I've played around with a spreadsheet solution as well using TMY3 weather data for all 8760 hours and columns for heat transfer/temperature/humidity across each device node in the process at each oa condition with fixed return/exhaust conditions.
> I will be very interested to hear/see if we get many responses to this. These units are a big portion of the energy consumption in k-12 schools and institutional occupancies in the warm and humid southeast. Thanks for kicking this discussion off.
> Dennis
> 
> On Thursday, March 21, 2013, Eric O'Neill wrote:
> 
> All,
> 
>  
> 
> I’m working on a retrocomissioning project trying to optimize a DOAS system as best as I can. This particular system has (in this order) an enthalpy wheel, preheating steam coil, cooling coil, sensible wheel, and heating coil. Downstream there are 4-pipe fan coil units. Hot water is provided by a steam to hot water converter and a chiller with a closed circuit tower that can be switched over to free cooling when ambient conditions allow.
> 
>  
> 
> I’ve made myself an excel model, which has worked to some extent, but when I want to ask some of the more challenging questions it’s going to require a not-so-modest commitment to developing it (and circular references are driving me nuts). I’m fairly confident I can’t do what I want to do with DOE2 (with which I’m most familiar). Is E+ capable of modeling to this level of detail? Some of the types of questions I’d like answers to:
> 
>  
> 
> ·         When it’s extremely hot outside and the enthalpy wheel and cooling coil are running to dehumidify, is it more efficient to run the sensible wheel (near) full speed (thus lowering the exhaust temperature, allowing some additional cooling from the enthalpy wheel, and lowering the cooling coil load) or to do more modest reheating to balance the downstream cooling and heating demands? My gut says it’s going to be best to run the sensible wheel only as fast as  is necessary to keep all the downstream heating coils closed, but I’d like some numbers.
> 
> ·         When temperatures are mild, but not sufficient for full water-side free cooling, what’s the right balance between the enthalpy wheel and terminal cooling and heating? Is the chiller operating at a part load that’s more cost effective than the heating, and how does the enthalpy wheel strike the right balance? Is there a sequence that is optimum, or would it be something that needs to be continually optimized by the controls system?
> 
>  
> 
> Obviously I’d need to go up the E+ learning curve, but I think that’s something I’ll need to do eventually (unless the fabled DOE2.3 engine comes out and can do this type of thing), so if it’s between that and continuing to develop a complicated excel model, it may be worth it for me. I just don’t want to go down that path only to find out I can’t get some of the answers I’m looking for…
> 
>  
> 
> Any thoughts (or references) are appreciated.
> 
>  
> 
> Cheers,
> 
> 
>  
>  
> 
> 
>  Eric L. O'Neill, P.E., LEED AP
> 
>  Managing Engineer - Implementation |  Michaels Energy
> 
>  Phone 608.785.3328 | Cell 608.792.7721
> 
>  www.michaelsenergy.com     
> 
> MichaelsNRG 
>  
> 
> 
> 
> -- 
> M. Dennis Knight, P.E.
> Founder & CEO
> Whole Building Systems, LLC
> P.O. Box 1845
> Mt. Pleasant, SC 29465
> Phone: 843-437-3647
> Email: dknight at wholebuildingsystems.com
> Website: www.wholebuildingsystems.com
> 
> 
> _______________________________________________
> Bldg-sim mailing list
> http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org
> To unsubscribe from this mailing list send  a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20130323/c79e8bfa/attachment-0001.htm>