<br>You can also look at a state-space model, I developed almost 20 years ago, that can directly output what you need. The model is now used in a tool DeST developed by Tsinghua University, China.<br><br>Tianzhen<br><br><div class="gmail_quote">
On Thu, Nov 1, 2012 at 1:00 PM, <span dir="ltr"><<a href="mailto:bldg-sim-request@lists.onebuilding.org" target="_blank">bldg-sim-request@lists.onebuilding.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Send Bldg-sim mailing list submissions to<br>
<a href="mailto:bldg-sim@lists.onebuilding.org">bldg-sim@lists.onebuilding.org</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org" target="_blank">http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org</a><br>
or, via email, send a message with subject or body 'help' to<br>
<a href="mailto:bldg-sim-request@lists.onebuilding.org">bldg-sim-request@lists.onebuilding.org</a><br>
<br>
You can reach the person managing the list at<br>
<a href="mailto:bldg-sim-owner@lists.onebuilding.org">bldg-sim-owner@lists.onebuilding.org</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of Bldg-sim digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. Re: Estimating system matrix A from simulation engine<br>
(Bryan Eisenhower)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Wed, 31 Oct 2012 17:24:40 -0700<br>
From: Bryan Eisenhower <<a href="mailto:bryane@engr.ucsb.edu">bryane@engr.ucsb.edu</a>><br>
To: <a href="mailto:bldg-sim@lists.onebuilding.org">bldg-sim@lists.onebuilding.org</a><br>
Cc: ??? <<a href="mailto:deukwookim@skku.edu">deukwookim@skku.edu</a>><br>
Subject: Re: [Bldg-sim] Estimating system matrix A from simulation<br>
engine<br>
Message-ID: <<a href="mailto:5091C148.6090501@engr.ucsb.edu">5091C148.6090501@engr.ucsb.edu</a>><br>
Content-Type: text/plain; charset="utf-8"; Format="flowed"<br>
<br>
Deuk-Woo,<br>
<br>
I have thought of this a lot before and will outline my experience<br>
below, in summary, if you are looking for accurate dynamical systems<br>
matrices from a building energy model (other than a hand-built RC<br>
network model), you will probably want to learn *Modelica*. Folks in the<br>
Modelica community have been using this method to study dynamics and<br>
controls for over a decade in the automotive, aerospace, and other<br>
industries.<br>
<br>
As for other simulation environments, its really up to the one you<br>
pick. Remember that the equations in an energy model are an<br>
approximation to reality. The numerical solution of these equations are<br>
yet another approximation to the dynamics in the equations. So, if you<br>
are going to do system identification on the model, you are going to get<br>
artifacts of both approximations. In my experience, the numerical<br>
solvers in programs like E+ and TRNSYS are not developed for accuracy in<br>
dynamics that you would need for this approach. They are developed more<br>
for long term analysis and numerical stability - both of which they do a<br>
great job on.<br>
<br>
I have been involved with a project to try and obtain ABCD matrices from<br>
*TRNSYS*. In this project, we went into the fortran code and wrote<br>
extra code to perturb state variables to observe how they relax and<br>
obtain matrices from this. We were successful in implementing this but<br>
did not trust the results because we felt the solver was overwriting the<br>
short term dynamics in the equations.<br>
<br>
I also worked on ABCD matrices using *EnergyPlus*, this time using a<br>
different approach. In this approach, I wrote an analytical model that<br>
can 'talk' to the EnergyPlus input file and .sql database files. In the<br>
sql file, there is most of the information about materials (e.g.<br>
capacitance), as well as all the bookkeeping on view factors and which<br>
constructions pertain to which zones. I wrote some scripts that read<br>
this file and automatically generate a linearized model. The linearized<br>
model does not take any time trace information (it does take some<br>
operating point information). It was a lot of work and although I trust<br>
the results, I wouldn't pass on the code to anyone else because it was a<br>
bit of a mess in the end. I have published this work this year and if<br>
you would like the papers, please write me personally.<br>
<br>
A fourth data point: *IDA-ICE* is another modeling tool that has a nice<br>
solver should capture short term dynamics. I asked Per Sahlin at IBPSA<br>
London this September if there was a way to produce ABCD matrices from<br>
the model and he said his customers were not interested (my paraphrasing<br>
- hopefully correct) - which is completely fair. My understanding is<br>
that the historical intersection of control analysis and building energy<br>
models has been on longer time scales (e.g. setback / setpoint analysis)<br>
or just time trace analysis - very little stability, frequency domain,<br>
or other advanced traditional techniques.<br>
<br>
Good luck! Since I have thought of this for some time, I'd be<br>
interested to discuss further with you, especially if you obtain any<br>
good results.<br>
<br>
Bryan<br>
<br>
<br>
<br>
On 10/30/2012 6:32 PM, ??? wrote:<br>
><br>
> Thank you for your reply.<br>
><br>
> Unfortunately, I?m not familiar with Modelica simulation environments.<br>
><br>
> If you don?t mind, could you please let me know any other ways?<br>
><br>
> Best Regards,<br>
> Deuk-Woo Kim<br>
><br>
> *From:*Michael Wetter [mailto:<a href="mailto:mwetter@lbl.gov">mwetter@lbl.gov</a>]<br>
> *Sent:* Wednesday, October 31, 2012 1:49 AM<br>
> *To:* ???<br>
> *Cc:* <a href="mailto:bldg-sim@lists.onebuilding.org">bldg-sim@lists.onebuilding.org</a><br>
> *Subject:* Re: [Bldg-sim] Estimating system matrix A from simulation<br>
> engine<br>
><br>
> Most Modelica simulation environments allow you to linearize a model<br>
> at the current state and save the A,B,C and D matrix that are its<br>
> state-space representation at the point of linearization. See<br>
><br>
> Michael Wetter.<br>
><br>
> Modelica-based Modeling and Simulation to Support Research and<br>
> Development in Building Energy and Control Systems.<br>
> <<a href="http://dx.doi.org/10.1080/19401490902818259" target="_blank">http://dx.doi.org/10.1080/19401490902818259</a>><br>
><br>
> /Journal of Building Performance Simulation/, 2(2):143-161, 2009.<br>
><br>
> Also available as LBNL Technical Report LBNL-2740E.pdf<br>
> <<a href="http://simulationresearch.lbl.gov/wetter/download/LBNL-2740E.pdf" target="_blank">http://simulationresearch.lbl.gov/wetter/download/LBNL-2740E.pdf</a>>.<br>
><br>
> where this is demonstrated for a nonlinear heat exchanger model. The<br>
> Modelica Buildings library<br>
> (<a href="http://simulationresearch.lbl.gov/modelica/" target="_blank">http://simulationresearch.lbl.gov/modelica/</a>) contains a room model<br>
> which you should be able to use for the same process to get the A,B,C<br>
> and D matrices for a building model.<br>
><br>
> Best,<br>
><br>
> Michael<br>
><br>
> On Sun, Oct 28, 2012 at 6:54 AM, ???<<a href="mailto:deukwookim@skku.edu">deukwookim@skku.edu</a><br>
> <mailto:<a href="mailto:deukwookim@skku.edu">deukwookim@skku.edu</a>>> wrote:<br>
><br>
> Dear all,<br>
><br>
> Is there a method to estimate system matrix *A* (of zone or building)<br>
> from simulation engines such as E+, ESP-r??<br>
><br>
> (The*A* is commonly found in state-space equation, e.g. x? = *A*x + Bu.)<br>
><br>
> As far as I know, only system identification method can be applied in<br>
> this situation.<br>
><br>
> (but, I not sure that it is reliable)<br>
><br>
> If there is more advanced way or efficient way to estimate *A, *please<br>
> let me know.<br>
><br>
> Any comments would be appreciated.<br>
><br>
> Thanks in advance.<br>
><br>
> Best Regards,<br>
> Deuk-Woo Kim<br>
><br>
><br>
> _______________________________________________<br>
> Bldg-sim mailing list<br>
> <a href="http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org" target="_blank">http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org</a><br>
> To unsubscribe from this mailing list send a blank message to<br>
> <a href="mailto:BLDG-SIM-UNSUBSCRIBE@ONEBUILDING.ORG">BLDG-SIM-UNSUBSCRIBE@ONEBUILDING.ORG</a><br>
> <mailto:<a href="mailto:BLDG-SIM-UNSUBSCRIBE@ONEBUILDING.ORG">BLDG-SIM-UNSUBSCRIBE@ONEBUILDING.ORG</a>><br>
><br>
><br>
><br>
> _______________________________________________<br>
> Bldg-sim mailing list<br>
> <a href="http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org" target="_blank">http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org</a><br>
> To unsubscribe from this mailing list send a blank message to <a href="mailto:BLDG-SIM-UNSUBSCRIBE@ONEBUILDING.ORG">BLDG-SIM-UNSUBSCRIBE@ONEBUILDING.ORG</a><br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <<a href="http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20121031/aabe0aef/attachment.html" target="_blank">http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20121031/aabe0aef/attachment.html</a>><br>
<br>
------------------------------<br>
<br>
_______________________________________________<br>
Bldg-sim mailing list<br>
<a href="mailto:Bldg-sim@lists.onebuilding.org">Bldg-sim@lists.onebuilding.org</a><br>
<a href="http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org" target="_blank">http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org</a><br>
<br>
<br>
End of Bldg-sim Digest, Vol 60, Issue 1<br>
***************************************<br>
</blockquote></div><br>