Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Problem with the convergence of Mooney Rivlin model
Posted Apr 15, 2020, 4:28 a.m. EDT Version 5.5 4 Replies
Please login with a confirmed email address before reporting spam
Dear all ; Im trying to model an hyperelastic tissue using Mooney Rivlin (MR) model in Comsol. When I use parameter values that satisfies the stability condition (below), the software never converges. This problem also occurs when I use two terms model. For all the examples that I could find and that work the C10 and C01 were positives. Did anyone face the same problem with Comsol? Stability conditions for MR: C10=-0.2948[MPa] // C10+C01>0 C01= 0.3179[MPa] C20=0.0300[MPa] /// C20+C02+C11>0 C11= 0.0715[MPa] C02 =-0.0973[MPa] /// C02<0 K = 0.4616[MPa] Regards
Please login with a confirmed email address before reporting spam
For analyses of this type, some solver settings may need changing. Are you running a stationary model? Are you apply a load or a displacement? How are you controlling the ramping of the load?
As a first step, try digging into the solver settings. Right click your Solver Configuration node and choose "show default solver" and under "Stationary Solver - Fully Coupled - Method and Termination" choose Constant (Newton). A little lower down in those same settings is an option to update the Jacobian, choose "on every iteration". A little further down, there is an option to choose a "Stablization and acceleration" setting. Select Anderson acceleration and increase the Dimensions of iteration space to 25.
Those settings have proven ideal for Hyperelastic MR and also Ogden defined materials.
Of course, the settings in the Physics nodes will also need scrutiny, but try the above first if you are confident in the settings you ave in your Physics nodes...
Please login with a confirmed email address before reporting spam
Thank you MortalWombatUk
Please login with a confirmed email address before reporting spam
Thank you MortalWombatUk
Please login with a confirmed email address before reporting spam
For analyses of this type, some solver settings may need changing. Are you running a stationary model? Are you apply a load or a displacement? How are you controlling the ramping of the load?
As a first step, try digging into the solver settings. Right click your Solver Configuration node and choose "show default solver" and under "Stationary Solver - Fully Coupled - Method and Termination" choose Constant (Newton). A little lower down in those same settings is an option to update the Jacobian, choose "on every iteration". A little further down, there is an option to choose a "Stablization and acceleration" setting. Select Anderson acceleration and increase the Dimensions of iteration space to 25.
Those settings have proven ideal for Hyperelastic MR and also Ogden defined materials.
Of course, the settings in the Physics nodes will also need scrutiny, but try the above first if you are confident in the settings you ave in your Physics nodes...
Hi, do you have a source where I can find more information on settings like these for hyperlastic material modeling? I'm also simulating with the Mooney-rivlin model, and can't get the solver to converge.