+- jeplus.org forums (http://jeplus.org/mybb)
+-- Forum: Building simulation tools (http://jeplus.org/mybb/forumdisplay.php?fid=1)
+--- Forum: jEPlus+EA (http://jeplus.org/mybb/forumdisplay.php?fid=3)
+--- Thread: using pre-calculated results (/showthread.php?tid=20)
Pages:
1
2
using pre-calculated results - kostas mourkos - 11-26-2014
Hi,
I went through the manual but I didn't undrstand how I can convert the simresults.csv in order to use it in jeplus+ea. Any help would be appreciated.
Thanks in advance,
Kostas
RE: using pre-calculated results - navid - 11-27-2014
Dear Kostas
If you wanna use jEPlus+EA,You have to have jEPlus project and Objective function.when you run jEPlus,AllCombinedResult.csv appears,if you click on action tab/view result/AllCombinedResults.csv [version:jEPlus_v1.5_beta_07]and then save it.the model simulation results are referenced as C0, C1, C2 and so on.Now you can make your Objective function and run jEplus+EA.
I hope,it helps you.
Best
Navid
RE: using pre-calculated results - navid - 11-27-2014
Dear Kostas
Everythings i have responded were for Local simulation,For more information,for Use pre-calculated results,The main difference between jEPlus result table ('SimResutls.csv') and pre-calculated result table file is in the 2nd column, the 'Job_Id'. jEPlus+EA has its [[[own naming convention]]] for solutions in a project. As a result, [[[manual conversion]]] from the jEPlus job IDs to the jEPlus+EA job IDs is [[necessary]].
Navid
RE: using pre-calculated results - kostas mourkos - 11-27-2014
Hi Navid,
Thanks for your repply.
Yes, that's my problem, I don't understand how to modify the 2nd column. For instance from a jeplus project that I have run I got in the 2nd column the foolowing:
EP_G-T_0-W_0-P1_0-P2_0-P3_0
EP_G-T_0-W_0-P1_0-P2_0-P3_2
EP_G-T_0-W_0-P1_0-P2_0-P3_1
EP_G-T_0-W_0-P1_0-P2_0-P3_3
So I am not sure about the conversion that I have to make.
Thanks again,
Kostas
RE: using pre-calculated results - navid - 11-27-2014
Dear Kostas
Honestly I Have not tried it.i Think it is too much bothersome for lots of variables.to modify the 2nd column,I think the following is your answer:
{ET}-{0}_{index of parameter 1(i)}_ {index of parameter 2(j)}_{index of parameter 3(k)}_ …
where [i,j,k,... are counters for your subvariables]and i=0:N j=0:M k=0:L ...
Where N,M,L,... are the Number of your variables.
e.g. parameter 1 is orientation[0:90:180],so:
ET-0_0 for 0
ET-0_1 for 90
ET-0_2 for 180
in jEPlus+EA file tool,in ET_PreCalc folder,There is .csv file called ET_10var_carbon_costs.csv. My above respond is from this file.maybe it helps you.
I Think above respond is the ansewr.But maybe Dear Dr Yi can answer better...
Navid
RE: using pre-calculated results - ivank - 11-27-2014
Dear Kostas,
I think navid is correct.
Since 'SimResults.csv' is a text file, you can open it with some text editor (such as NotePad++) and 'Find&Replace' extra characters ('-P01', etc.)
Best,
Ivan
RE: using pre-calculated results - kostas mourkos - 11-28-2014
Hi guys,
I see now how it works, it is not so easy as I thought.
Thanks for the help,
Kostas
RE: using pre-calculated results - Yi - 11-28-2014
Hi Kostas, Navid, sorry for coming in late ...
The pre-calculated result option is for demonstrating how optimisation works only. Everything in the table will be plotted as the gray cloud in the background, so you can see the whole search space and how the algorithm explores the solutions.
The format of the result table is like this:
- the first row (headers) is ignored
- column 1 is index of the rows
- column 2 is the job ids. These are jEPlus+EA job ids, which are slightly different to the jEPlus's. It starts with the project initial which is set in the .jep file. The rest part is indexes of the parameter values used in each case, separated by an underscore ('_'). Note that first index is for the IDF model. Weather file cannot be a design variable so it is not represented here. In the ET_PreCalc example, you see 11 digits in the job ids, representing the IDF model and the 10 jEPlus parameters.
- column 3 is ignored
- columns 4, 5 ... are objective function values, coming from the full parametric run or other sources.
Hope this will make it a bit clearer.
By the way, Kostas, why do you want to use this function?
Cheers,
Yi
RE: using pre-calculated results - kostas mourkos - 11-29-2014
Hi Yi,
Basically I would like to add an additional objective, daylight autonomy obtained from Daysim.
Kostas
RE: using pre-calculated results - Yi - 11-29-2014
Ah, I see. The pre-calculated table format is correct for this purpose. However, to use the data as objective functions, you need to specify it in the .rvi file. Below is the 5ZoneCostEst.rvi file in the example_2-rvx_E+v8.1\ folder in the jEPlus package:
Code:
eplusout.mtr
eplusout.csv
Electricity:Facility
0
!-sqlite
! Output file name; Column headers; SQL command
ChillerCap; Chiller Nominal Capacity
ConsCost; Construction Cost
!-end sqlite
!-user spreadsheet
! Output file name; User spreadsheet file name; columns to extract (e.g. 3,5, 6, 7)
!-end user spreadsheet
!-objectives
! Name of the objective; unit; formula (c0,c1, c2... are columns in the result table generated from the statements above)
Electricity; kWh; c0/1000/3600
Chiller Capacity; kW; c1/1000
Construction Cost; $/m2; c2
!-end objectivesYou will need to fill in the !-user spreadsheet block with the relevant information. The "columns to extract" list should start from '3', which is the fourth column in the pre-calc table.
I can imagine if you have a large number of cases, this file will be quite difficult to compile. This is why we are ditching it for a more flexible Python solution, which will come with the next release of jEPlus+EA. For the time being, though, there is another (undocumented) trick you can use. The job_ids in the table can be regular expressions, which means you can use one row to cover multiple cases.
For example if you have 6 parameters (including the IDF model), a job_id should look like EP-0_0_0_0_0_0. If only parameter 2 and 3 make difference to your daysim results, you can write rows like these:
Code:
Id,Job_Id,reserved,
0,EP-[0-9]+_0_0_[0-9]+_[0-9]+_[0-9]+,-,2234.8
1,EP-[0-9]+_0_1_[0-9]+_[0-9]+_[0-9]+,-,2101.8
2,EP-[0-9]+_0_2_[0-9]+_[0-9]+_[0-9]+,-,2058.6
3,EP-[0-9]+_1_0_[0-9]+_[0-9]+_[0-9]+,-,2188.7
...So the first row assign 2234.8 to all cases whose job_ids have 0 and 0 for the 2nd and the 3rd parameters, respectively. If you haven't used regular expression before, this can be a bit of learning curve; but it is a very powerful tools.
Yi