Bone remodelling after Total Hip Replacement

[nazman]

hi, my name is nazman, from malaysia. rite now i m doing my master degree in universiti teknologi malaysia. and, basically, i got the same problem with raptor22 which has wrote like this before;

Hi all:

I am trying to analyse a problem. The problem description is:

I have an FE model, the model represents a human bony structure. The Young's modulus/density of the structure will change according to the mechanical stresses in the bone structure.

I am using Abaqus 6.6-1 and Fortran95.

Therefore in short. What I need to do is to link Fortran 95 to Abaqus, so when I run the FE model, the material property of the model can be changed automatically by Fortran continously, simulating real time bone remodeling (bone is a dynamic substance)

To analyse this problem, I thought I could use Abaqus to calculate the stress, therefore strain energy density of the FE model. Use the resulting strain energy values from the FE calculation, then put those values into a Fortran 95 user subroutine program, calculate the new Young's modulus values and update the material properties in the abaqus input file.

At the moment, I think I will have to somehow output an energy file from Abaqus, and import the values into the Fortran program. Then use Fortran to calculate the new Young's modulus values, then export the values from Fortran95 back to abaqus input file again.

Is that right? Or is there a better way of doing it?

Any help will be greatly appreciated!

Therefore, can anyone, mind to help me? or, maybe someone can help me to track raptor22, or give me the email address of this guy (raptor22)? maybe he has already solve that problem rite now and i can liase directly with him..

thanx a lot frenz..... :D

[ashu28]

I would suggest store energy and modulus in state variables, write updation algorithm, and then calculate stresses.

Using state variables to your advantage would be the key to the problem.

Ashu

[Jorgen]

The policy of the PolymerFEM.com forum is not to give away private email address of members. If you want to reach raptor22 (or some other member), then perform the following steps:

(1) Go to the following page and search for the user name.
(2) After you have found the user name you can then send a private message directory to that member.

- Jorgen

[microfem]

Hi

I am also working on a project where I have to simulate bone remodelling using abaqus and UMAT.

In my input file I start with a lower young's modulus and this is the value UMAT takes at the first iteration to calculate stresses.

Then it calculates the new young's modulus using the following equation.

EMOD=EMOD+EMOD*B*(((stran(1)+stran(2))/2)-control stran)

this is the DeltaEMOD = EMOD*B*(((stran(1)+stran(2))/2)-control stran)

B is the rate of bone adaptation

STATEV(1)=EMOD

Then it saves this as a state variable and pass it on to input file, where one has to use *DEPVAR in order to accept this value and at next iteration this is given back to the UMAT to continue with the stresses calculation and material propeties calculation.

My problem is how should I ensure that the program terminates when the EMOD is equal to the final EMOD as there is no error signal from the above control equation any more. I can use a condition here to observe whether the average strain is equal to expected strain or whether EMODS are equal and that EMOD should stay the same, but how do I know that the average strain would not change further having reached the expected strain, in other words that there will be no more bone remodelling. (It is better if the operation can terminate having converged)

Any comment and advice on this regard would be greatly appreciated.