Mixing
In Depth Treatment of 1-D Adaptive Mesh Refinement Analytics and Implementation
Clash Royale CLAN TAG#URR8PPP
up vote
0
down vote
favorite
I am looking for a good resource for implementing adaptive mesh refinement (AMR) on a 1-D PDE and an analytical treatment. Pretty much everything I've found on AMR use 1-D as a toy example to show the benefits and then goes straight to 2-D problems, which are much harder mathematically and computationally for several reasons.
For a little background on why I would possibly need AMR for a 1-D problem, I want to solve the Cable equation, which models voltages in neurons. The Cable equation is the heat equation with a forcing term, and it is solved on a network of connected 1-D rods. (If you think of the typical artists rendition of a neuron, it is a branching network of cylindrical tubes.) Simulations are time consuming because there are usually a lot compartments in a neuron and a lot of neurons in a network, so a 50% speed up is a big deal.
I think AMR could really speed up the typical algorithm because in this paper, Rempe MJ, Spruston N, Kath WL, Chopp DL. Compartmental Neural Simulations with Spatial Adaptivity, they got around a 50% decrease in computation time simply by not simulating the parts of the neuron at steady state, no adaptive mesh. So if I used their idea along with an adaptive mesh, the decrease in computation time will be even better than 50%, which would be awesome!
So if anyone knows of a rigorous treatment of AMR in 1-D, along with good data structures to handle the changing 1-D mesh, please let me know. (I don't want to reinvent the wheel if I don't have to.) Also, I have a BS in computer science and an MS in applied math, so nothing in 1-D will go over my head.
Edit:
I was able to find this paper, [An adaptive mesh-moving and refinement procedure for one-dimensional conservation laws][1], but it was published in 1993. The paper has the analytical work I wanted, and a dynamic tree was used to handle the adaptive mesh.
I checked the citing papers as well, and there doesn't seem to be much work on 1-D adaptive mesh refinement. Maybe 1-D is too easy to justify much research, but I was hoping there was a more in depth look into which data structures work well with certain problems.
I will start digging in and answer this question once I have a satisfactory program.
numerical-methods biology data-structure
asked 2 days ago
Mitchal
12
add a comment |Â
up vote
0
down vote
favorite
I am looking for a good resource for implementing adaptive mesh refinement (AMR) on a 1-D PDE and an analytical treatment. Pretty much everything I've found on AMR use 1-D as a toy example to show the benefits and then goes straight to 2-D problems, which are much harder mathematically and computationally for several reasons.
For a little background on why I would possibly need AMR for a 1-D problem, I want to solve the Cable equation, which models voltages in neurons. The Cable equation is the heat equation with a forcing term, and it is solved on a network of connected 1-D rods. (If you think of the typical artists rendition of a neuron, it is a branching network of cylindrical tubes.) Simulations are time consuming because there are usually a lot compartments in a neuron and a lot of neurons in a network, so a 50% speed up is a big deal.
I think AMR could really speed up the typical algorithm because in this paper, Rempe MJ, Spruston N, Kath WL, Chopp DL. Compartmental Neural Simulations with Spatial Adaptivity, they got around a 50% decrease in computation time simply by not simulating the parts of the neuron at steady state, no adaptive mesh. So if I used their idea along with an adaptive mesh, the decrease in computation time will be even better than 50%, which would be awesome!
So if anyone knows of a rigorous treatment of AMR in 1-D, along with good data structures to handle the changing 1-D mesh, please let me know. (I don't want to reinvent the wheel if I don't have to.) Also, I have a BS in computer science and an MS in applied math, so nothing in 1-D will go over my head.
Edit:
I was able to find this paper, [An adaptive mesh-moving and refinement procedure for one-dimensional conservation laws][1], but it was published in 1993. The paper has the analytical work I wanted, and a dynamic tree was used to handle the adaptive mesh.
I checked the citing papers as well, and there doesn't seem to be much work on 1-D adaptive mesh refinement. Maybe 1-D is too easy to justify much research, but I was hoping there was a more in depth look into which data structures work well with certain problems.
I will start digging in and answer this question once I have a satisfactory program.
numerical-methods biology data-structure
asked 2 days ago
Mitchal
12
If you already have an understanding of how AMR is applied to a toy problem, what is the difficulty with applying the same thing to a non-toy problem? Can you be more specific about the challenges you are facing?
– Rahul
8 hours ago
add a comment |Â
up vote
0
down vote
favorite
up vote
0
down vote
favorite
I am looking for a good resource for implementing adaptive mesh refinement (AMR) on a 1-D PDE and an analytical treatment. Pretty much everything I've found on AMR use 1-D as a toy example to show the benefits and then goes straight to 2-D problems, which are much harder mathematically and computationally for several reasons.
For a little background on why I would possibly need AMR for a 1-D problem, I want to solve the Cable equation, which models voltages in neurons. The Cable equation is the heat equation with a forcing term, and it is solved on a network of connected 1-D rods. (If you think of the typical artists rendition of a neuron, it is a branching network of cylindrical tubes.) Simulations are time consuming because there are usually a lot compartments in a neuron and a lot of neurons in a network, so a 50% speed up is a big deal.
I think AMR could really speed up the typical algorithm because in this paper, Rempe MJ, Spruston N, Kath WL, Chopp DL. Compartmental Neural Simulations with Spatial Adaptivity, they got around a 50% decrease in computation time simply by not simulating the parts of the neuron at steady state, no adaptive mesh. So if I used their idea along with an adaptive mesh, the decrease in computation time will be even better than 50%, which would be awesome!
So if anyone knows of a rigorous treatment of AMR in 1-D, along with good data structures to handle the changing 1-D mesh, please let me know. (I don't want to reinvent the wheel if I don't have to.) Also, I have a BS in computer science and an MS in applied math, so nothing in 1-D will go over my head.
Edit:
I was able to find this paper, [An adaptive mesh-moving and refinement procedure for one-dimensional conservation laws][1], but it was published in 1993. The paper has the analytical work I wanted, and a dynamic tree was used to handle the adaptive mesh.
I checked the citing papers as well, and there doesn't seem to be much work on 1-D adaptive mesh refinement. Maybe 1-D is too easy to justify much research, but I was hoping there was a more in depth look into which data structures work well with certain problems.
I will start digging in and answer this question once I have a satisfactory program.
numerical-methods biology data-structure
asked 2 days ago
Mitchal
12
I am looking for a good resource for implementing adaptive mesh refinement (AMR) on a 1-D PDE and an analytical treatment. Pretty much everything I've found on AMR use 1-D as a toy example to show the benefits and then goes straight to 2-D problems, which are much harder mathematically and computationally for several reasons.
For a little background on why I would possibly need AMR for a 1-D problem, I want to solve the Cable equation, which models voltages in neurons. The Cable equation is the heat equation with a forcing term, and it is solved on a network of connected 1-D rods. (If you think of the typical artists rendition of a neuron, it is a branching network of cylindrical tubes.) Simulations are time consuming because there are usually a lot compartments in a neuron and a lot of neurons in a network, so a 50% speed up is a big deal.
I think AMR could really speed up the typical algorithm because in this paper, Rempe MJ, Spruston N, Kath WL, Chopp DL. Compartmental Neural Simulations with Spatial Adaptivity, they got around a 50% decrease in computation time simply by not simulating the parts of the neuron at steady state, no adaptive mesh. So if I used their idea along with an adaptive mesh, the decrease in computation time will be even better than 50%, which would be awesome!
So if anyone knows of a rigorous treatment of AMR in 1-D, along with good data structures to handle the changing 1-D mesh, please let me know. (I don't want to reinvent the wheel if I don't have to.) Also, I have a BS in computer science and an MS in applied math, so nothing in 1-D will go over my head.
Edit:
I was able to find this paper, [An adaptive mesh-moving and refinement procedure for one-dimensional conservation laws][1], but it was published in 1993. The paper has the analytical work I wanted, and a dynamic tree was used to handle the adaptive mesh.
I checked the citing papers as well, and there doesn't seem to be much work on 1-D adaptive mesh refinement. Maybe 1-D is too easy to justify much research, but I was hoping there was a more in depth look into which data structures work well with certain problems.
I will start digging in and answer this question once I have a satisfactory program.
numerical-methods biology data-structure
asked 2 days ago
Mitchal
12
edited 8 hours ago
asked 2 days ago
Mitchal
12
asked 2 days ago
Mitchal
12
asked 2 days ago
Mitchal
12
12
If you already have an understanding of how AMR is applied to a toy problem, what is the difficulty with applying the same thing to a non-toy problem? Can you be more specific about the challenges you are facing?
– Rahul
8 hours ago
add a comment |Â
If you already have an understanding of how AMR is applied to a toy problem, what is the difficulty with applying the same thing to a non-toy problem? Can you be more specific about the challenges you are facing?
– Rahul
8 hours ago
If you already have an understanding of how AMR is applied to a toy problem, what is the difficulty with applying the same thing to a non-toy problem? Can you be more specific about the challenges you are facing?
– Rahul
8 hours ago
If you already have an understanding of how AMR is applied to a toy problem, what is the difficulty with applying the same thing to a non-toy problem? Can you be more specific about the challenges you are facing?
– Rahul
8 hours ago
add a comment |Â
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2871743%2fin-depth-treatment-of-1-d-adaptive-mesh-refinement-analytics-and-implementation%23new-answer', 'question_page');
);
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
If you already have an understanding of how AMR is applied to a toy problem, what is the difficulty with applying the same thing to a non-toy problem? Can you be more specific about the challenges you are facing?
– Rahul
8 hours ago