Discussion Forum

Hello all,

I am trying to model the operation of neural dust, as described by Dongjin Seo et al: Neural Dust

I've set up my model with 3 components: ultrasonic transmitter, propagating medium and simplified neural dust mote, and ultrasonic reciever. I am basing my transmitter and receiver off of the COMSOL tutorial for the ultrasonic flowmeter. The tutorial points out that to couple ultrasound with the piezoelectric transducers they had to use separate studies as a workaround. The reason they had to do this is because the convected wave equation is modeled by the DG method, while the piezoelectric is modeled by the FEM method, and the two cannot be directly coupled. The separation into several studies is fine for the transmitter and receiver interfaces, and honestly preferable because it descretizes my individual study steps. However, at the mote reflection interface I'm not sure that I can separate the two studies, because the reflections depend directly on the piezoelectric characteristics of the mote at the time the ultrasound hits. I was wondering if the COMSOL community had any insight into how I could work around this. Could I manually make both models work in the DG method? Or is there another way I can model reflections? Any advice you may have would be very helpful.

Thanks, Stephen