Discussion Forum

Hi, I am building a bioheat transfer model. Generally speaking, I want to control the temperature change in human tissue by varying the power dissipation on microelectronics. I have already set the initial condition in COMSOL.

Currently I am working on using Matlab to return the current time step temperature at a specific point (will not change during the simulation) in the model and calculate the power dissipation for the next time step. If the temperature does not exceed the thermal limit (constant value), the the power dissipation for the next time step is based only the current temperature; if the temperature exceeds thermal limit, the power dissipation will be forced to zero during the next time step.

The problem is that I don't know how to run the simulation in COMSOL for just one time step, return the temperature to Matlab, do the calculation in Matlab and then resume the simulation for the next time step with the newly calculated power dissipation. Is it possible to do this procedure? If so, how can I improve the time efficiency?

Plus, another method that has come up my mind is to set a probe in COMSOL and track the temperature for each time step. Meanwhile, I will set the power dissipation as a variable (perhaps a global variable?) in the form of an expression so that I can update its value based on current temperature. In this way I don't need to switch between Matlab and COMSOL. However, it seems that COMSOL does not like complicated formula expressions. What's more, I have no idea if COMSOL will update the variable at the end of each time step.

Using the power management algorithm on this thermal model is the key topic of my team's research. I hope with the help of experienced COMSOL users I can figure it out ASAP.