Plug Flow Reactor MATLAB code for changing are and straight area #701

mgashwinkumar · 2019-08-20T12:30:53Z

Please fill in the issue number this pull request is fixing:

Fixes #

Changes proposed in this pull request:

codecov · 2019-08-20T13:00:15Z

Codecov Report

Merging #701 into master will increase coverage by 0.17%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #701      +/-   ##
==========================================
+ Coverage   71.40%   71.57%   +0.17%     
==========================================
  Files         372      372              
  Lines       43605    44499     +894     
==========================================
+ Hits        31134    31849     +715     
- Misses      12471    12650     +179

Impacted Files	Coverage Δ
include/cantera/zeroD/IdealGasReactor.h	`50.00% <0.00%> (-50.00%)`	⬇️
include/cantera/zeroD/ConstPressureReactor.h	`50.00% <0.00%> (-50.00%)`	⬇️
...clude/cantera/zeroD/IdealGasConstPressureReactor.h	`50.00% <0.00%> (-50.00%)`	⬇️
include/cantera/kinetics/Falloff.h	`52.50% <0.00%> (-47.50%)`	⬇️
include/cantera/kinetics/Kinetics.h	`56.75% <0.00%> (-33.49%)`	⬇️
include/cantera/numerics/ctlapack.h	`77.04% <0.00%> (-22.96%)`	⬇️
include/cantera/thermo/MixtureFugacityTP.h	`20.00% <0.00%> (-13.34%)`	⬇️
include/cantera/thermo/SingleSpeciesTP.h	`54.54% <0.00%> (-12.13%)`	⬇️
src/base/units.h	`93.04% <0.00%> (-6.96%)`	⬇️
include/cantera/thermo/NasaPoly2.h	`96.00% <0.00%> (-4.00%)`	⬇️
... and 55 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b23f184...e9bd282. Read the comment docs.

mgashwinkumar · 2019-09-06T00:01:37Z

What is the status of this pull request. Is there any comments? Is there anything that I should look into?

bryanwweber · 2019-09-06T11:25:44Z

Hi! Sorry that no one has responded to you. We haven't had the time to review this yet. Thanks for your patience!

bryanwweber

Hello! Thank you again for this contribution. I have a few concerns broadly about this example, in addition to the specific comments listed below. Please feel free to ask any questions or push back on any suggestion.

Can you provide a short source/description with the derivation of the equations you're using here, particularly the conservation equations for mass, energy, and species?
You are using the equilibrated gas_conv object passed into all your calls. This results in the composition being essentially fixed and nothing much interesting happening (i.e., this could be solved by frozen composition calculations without an ODE I think). Was this intentional? I tried to change the state to the equilibrated temperature, but the initial composition, but the simulation took a long time and I didn't wait for it to finish.
The files that you're adding here should be placed in the samples/matlab folder, please.

Thanks again for your contribution, and please do let me know any questions you have!

bryanwweber · 2019-09-28T17:20:32Z

General_Plug_Flow_Reactor.m

@@ -0,0 +1,75 @@
+% This code snippet is a sample to model a converging nozzle as


The documentation for this example should be in the format similar to flame1.m in the samples/matlab folder. In other words, something like

% GENERAL_PLUG_FLOW_REACTOR - One line description % % Longer description goes here indented by four spaces after the % comment character

bryanwweber · 2019-09-28T17:37:14Z

General_Plug_Flow_Reactor.m

+in2  = speciesIndex(gas_conv,'N2');
+nsp = nSpecies(gas_conv);
+x = zeros(nsp,1);
+% Change the below values for different Phi values of methane Combsution


A small typo here... "Combsution" -> "combustion"

bryanwweber · 2019-09-28T17:37:47Z

General_Plug_Flow_Reactor.m

+nsp = nSpecies(gas_conv);
+x = zeros(nsp,1);
+% Change the below values for different Phi values of methane Combsution
+x(ich4,1) =0.2899; 


Can you make sure the spacing before and after the = sign is consistent with the rest of the code? Also, can you delete the space at the end of the line?

bryanwweber · 2019-09-28T17:38:23Z

General_Plug_Flow_Reactor.m

+
+%% Converging Section
+% The Dimensions and conditions of the reactor are given below
+A_in=0.018;            % Inlet Area


I would prefer if the different quantities were labeled by a comment above the variable, rather than on the same line.

bryanwweber · 2019-09-28T17:38:43Z

General_Plug_Flow_Reactor.m

+% k = -1 makes the solver solve for converging area. 
+% Use k = +1 for diverging area. make sure to change the inlet and exit area for diverging area
+% Use k = 0 for constant cross sectional area
+k=-1;     


Can you automatically set k based on the relative values for A_in and A_out?

bryanwweber · 2019-09-30T18:59:19Z

PFR_solver_CANTERA.m

+%---density, temperature and mass fractions variations along a plug flow---
+%-------------------------reactor------------------------------------------
+%--------------------------------------------------------------------------
+F(1)=((1-R/cp_mass(gas))*((rho*vx)^2)*(1/A)*(dAdx) + rho*R*sum(MW.*w.*(h-MW_mix*cp_mass(gas)*T./MW))/(vx*cp_mass(gas)) ) / (P*(1+vx^2/(cp_mass(gas)*T)) - rho*vx^2);


You may want to add a variable for the value of cp_mass to avoid having to calculate it so many times in these two lines.

I think you should also split this calculation over several lines by defining several variables to hold different pieces of the equation. That will make it easier for people to see how each term is calculated.

It is easier for me like this to debug. If everything else is approved, I can change this as the last step before you upload it to the main repository

bryanwweber · 2019-09-30T19:01:46Z

PFR_solver_CANTERA.m

+if k~=0
+    set(gas,'Temperature',T,'Density',rho,'MassFractions',Y); % the gas is set to the corresponding properties during each iteration of the ode loop
+else
+    set(gas,'Temperature',T,'Pressure',P,'MassFractions',Y); % the gas is set to the corresponding properties during each iteration of the ode loop


Why use pressure to set the state when k = 0? Density should be equally as valid, right?

Yes. I changed it

bryanwweber · 2019-09-30T19:02:45Z

PFR_solver_CANTERA.m

@@ -0,0 +1,43 @@
+function [F] = PFR_solver_CANTERA(x,initial,gas,mdot,A_in,dAdx,k)


Instead of naming this function PFR_solver_CANTERA, can you just name it PFR_solver? I think you should also be able to include it in the PFR_setup.m file, at the end of that file. I'm not that familiar with how MATLAB functions are defined though.

bryanwweber · 2019-09-30T19:03:47Z

PFR_solver_CANTERA.m

+
+if k==1
+    A = A_in-k*dAdx*x;
+elseif k==-1|k==0


My linter suggests using || for this comparison. However, if k = 1, doesn't this result in the area decreasing anyways? I think you have a logic error here.

Good catch. I fixed it

bryanwweber · 2019-09-30T19:10:07Z

PFR_solver_CANTERA.m

+F(2) = (vx*vx/(rho*cp_mass(gas)))*F(1) + vx*vx*(1/A)*(dAdx)/cp_mass(gas) - (1/(vx*rho*cp_mass(gas)))*sum(h.*w.*MW);
+
+for i = 3:nsp+2
+    F(i) = w(i-2)*MW(i-2)/(rho*vx);


Instead of a for-loop, can you do this calculation all at once, and then add the calculated values into the F array?

bryanwweber · 2019-12-03T02:35:35Z

@mgashwinkumar Hi 👋 Do you have any questions about the comments that I left here? How to update the code, or anything like that?

mgashwinkumar · 2019-12-03T02:36:59Z

Bryan, Thanks for reaching out. I didn’t get a chance to review the questions you posted to me. I will do that and upload the revised version soon. Regards Ashwin From: Bryan W. Weber Sent: Monday, December 2, 2019 9:35 PM To: Cantera/cantera Cc: mgashwinkumar; Mention Subject: Re: [Cantera/cantera] Plug Flow Reactor MATLAB code for changing areand straight area (#701) @mgashwinkumar Hi 👋 Do you have any questions about the comments that I left here? How to update the code, or anything like that? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

bryanwweber · 2020-05-29T19:03:48Z

@mgashwinkumar Hello again! Just wondering when/if you'll have a chance to update this code? Thanks!

mgashwinkumar · 2020-05-31T14:15:27Z

@bryanwweber , I will update the code and try to address your comments in two-three weeks.

bryanwweber · 2020-05-31T14:16:37Z

@mgashwinkumar thanks! Let me know if you have any questions, I'm happy to help

mgashwinkumar · 2020-05-31T18:11:33Z

@mgashwinkumar thanks! Let me know if you have any questions, I'm happy to help

@bryanwweber , Is there a way to retrieve the version that I uploaded earlier for reference?

bryanwweber · 2020-06-01T12:03:28Z

@mgashwinkumar You can go to https://github.com/Cantera/cantera/pull/701/files

and then click the 3 dots, and then "View File" which will let you see the original version.

mgashwinkumar · 2020-06-02T00:47:50Z

@bryanwweber , I replied to each of your comment individually. I have uploaded the derivations in a pdf along with this comment. I have also uploaded the new files into samples/matlab of my individual repository. Please let me know if you have any other question or concern.
PFR Derivation.pdf

bryanwweber · 2020-06-02T00:51:32Z

@mgashwinkumar Thank you! Unfortunately, I can't see any of the changes here. Did you do them on your local computer? If so, you'll need to git commit them, and then git push them so they're on GitHub. Let me know if you need any help with that!

mgashwinkumar · 2020-06-02T01:13:44Z

Bryan, Yes. I changed it on my local Matlab. I just uploaded it. I didn’t do any git commit. Can you please help with me. Also, if you don’t mind, are you available to meet over zoom sometime tomorrow so that you can show me how to do and I don’t have to bother you every time. Sorry, its just my lack of experience with GitHub. RegardsAshwin From: Bryan W. WeberSent: Monday, June 1, 2020 8:51 PMTo: Cantera/canteraCc: mgashwinkumar; MentionSubject: Re: [Cantera/cantera] Plug Flow Reactor MATLAB code for changing are and straight area (#701) @mgashwinkumar Thank you! Unfortunately, I can't see any of the changes here. Did you do them on your local computer? If so, you'll need to git commit them, and then git push them so they're on GitHub. Let me know if you need any help with that!—You are receiving this because you were mentioned.Reply to this email directly, view it on GitHub, or unsubscribe.

bryanwweber · 2020-06-02T02:00:44Z

@mgashwinkumar No problem! Send me an email to the email on my profile and we can set up a meeting. Tomorrow won't work for me, but Wednesday or Thursday will be a little better.

bryanwweber · 2020-06-02T13:32:24Z

@mgashwinkumar Wednesday at 10 would work for me. Can you send a separate Zoom invite by email? I don't think it's a good idea to publish meeting codes publicly like this.

bryanwweber · 2020-06-03T15:36:45Z

@mgashwinkumar I made a few more fixes to the files here, let me know if anything seems wrong! https://github.com/Cantera/cantera/pull/701/files

mgashwinkumar · 2020-06-03T16:16:06Z

@bryanwweber , I would like to download the latest version and run it once on my computer to check if everything is fine. Should I download the file from the main cantera repository or the forked copy of mine?

bryanwweber · 2020-06-03T17:09:45Z

@mgashwinkumar the forked copy, all the changes got made to those copies

decaluwe · 2020-06-03T21:10:34Z

Hi, @mgashwinkumar -- thanks so much for submitting this PR - I think it will be well-received and useful for many!

I made a few small changes, mostly to documentation and variable names in a couple places, but otherwise this should be good to go!

mgashwinkumar · 2020-06-04T00:01:52Z

@bryanwweber , @decaluwe , Thank you very much for your inputs, I have a few comments/questions mentioned below:

I tried running the code, my matlab was having trouble with '.yam1'. When I changed it to '.cti', it ran smoothly. I am not sure why this is happening and if this will affect someone who is trying to use this in matlab. If that isn't a problem, I'm fine with this version. What is the next step?
I just added contact info at the end of description file if that's fine.
Just curious, if this is the final version, does it show up on the Cantera Matlab examples webpage ?

mgashwinkumar · 2020-06-04T22:03:25Z

@bryanwweber , Any updates about the final version and the comment I posted yesterday?

The reactor can have varying area, and therefore, pressure variation throughout the domain. The example is solved by discretizing the domain into chunks and solving them as a sequence of 0-D reactors.

Renaming solution vector references. Renaming 'init' variable, external and internal to the residual function for greater clarity. External to the ode call, they are renamed 'inlet_soln', since they correspond to the current volume's inlet state. Within the ode call, they are renamed 'soln_vector' since they represent solution at the current location x.

The call to meanMolecularWeight was previously located before the Cantera gas object was set to the current state. This commit fixes that.

bryanwweber · 2020-06-05T01:22:21Z

@mgashwinkumar No problem adding the contact info. The YAML file is something new in Cantera 2.5.0, so I wouldn't expect it to work in your environment unless you compiled Cantera yourself fairly recently. Using the CTI version is fine for your local testing, but since this will go out with Cantera 2.5.0, we'd like it to use the new format. Once we release v2.5.0 (sometime this summer), your example will show up on the website! Thanks again for the contribution, I'm just going to wait for the tests to pass before merging it into master 😃

mgashwinkumar · 2020-06-05T01:37:39Z

@bryanwweber , Thank you very much for your help. I am happy that I contributed to the combustion community. Please let me know when you publish it online.

mgashwinkumar changed the title ~~Add files via upload~~ Plug Flow Reactor MATLAB code for changing are and straight area Aug 20, 2019

bryanwweber requested changes Sep 30, 2019

View reviewed changes

bryanwweber force-pushed the mgashwinkumar-patch-2 branch from 762c776 to 5883fd8 Compare June 3, 2020 14:22

mgashwinkumar and others added 5 commits June 4, 2020 21:12

Add plug flow reactor example in MATLAB

aa9c9a5

The reactor can have varying area, and therefore, pressure variation throughout the domain. The example is solved by discretizing the domain into chunks and solving them as a sequence of 0-D reactors.

Renaming Matlab plug flow reactor example file.

0942ebd

Moving PFR_solver call to 'meanMolecularWeight'

30fd0ce

The call to meanMolecularWeight was previously located before the Cantera gas object was set to the current state. This commit fixes that.

Add author contact information to sample

e9bd282

bryanwweber force-pushed the mgashwinkumar-patch-2 branch from 2588556 to e9bd282 Compare June 5, 2020 01:20

bryanwweber approved these changes Jun 5, 2020

View reviewed changes

bryanwweber merged commit 57b4641 into Cantera:master Jun 5, 2020

		@@ -0,0 +1,75 @@
		% This code snippet is a sample to model a converging nozzle as

		@@ -0,0 +1,43 @@
		function [F] = PFR_solver_CANTERA(x,initial,gas,mdot,A_in,dAdx,k)

Plug Flow Reactor MATLAB code for changing are and straight area #701

Plug Flow Reactor MATLAB code for changing are and straight area #701

Conversation

mgashwinkumar commented Aug 20, 2019

Changes proposed in this pull request:

codecov bot commented Aug 20, 2019 • edited Loading

Codecov Report

mgashwinkumar commented Sep 6, 2019

bryanwweber commented Sep 6, 2019

bryanwweber left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

bryanwweber commented Dec 3, 2019

mgashwinkumar commented Dec 3, 2019 via email

bryanwweber commented May 29, 2020

mgashwinkumar commented May 31, 2020

bryanwweber commented May 31, 2020

mgashwinkumar commented May 31, 2020

bryanwweber commented Jun 1, 2020

mgashwinkumar commented Jun 2, 2020

bryanwweber commented Jun 2, 2020

mgashwinkumar commented Jun 2, 2020 via email

bryanwweber commented Jun 2, 2020

bryanwweber commented Jun 2, 2020

bryanwweber commented Jun 3, 2020

mgashwinkumar commented Jun 3, 2020

bryanwweber commented Jun 3, 2020

decaluwe commented Jun 3, 2020

mgashwinkumar commented Jun 4, 2020

mgashwinkumar commented Jun 4, 2020

bryanwweber commented Jun 5, 2020

mgashwinkumar commented Jun 5, 2020

codecov bot commented Aug 20, 2019 •

edited

Loading