Using the Optimization Lab in the classroom
An inverse model created with the help of the Optimization Lab determines the critical spinning frequency at which a gear separates from its shaft.
Accessed from the COMSOL Script command line, this simulation tool includes powerful optimization solvers based on the highly respected and widely used SNOPT and SQOPT codes developed by Philip Gill from the University of California at San Diego, along with Walter Murray and Michael Saunders from Stanford University. It includes:
- Ease of use—Students access optimization functions through an interactive programming interface. Here they provide a vector of parameters and a set of arbitrary constraints along with a single quantity to optimize
- Constrained problems—Linear, quadratic, and nonlinear constrained problems can be optimized along with linear and nonlinear least-squares constrained problems
- Unconstrained problems—It contains the Nelder-Mead search algorithm for unconstrained nonlinear optimization, which is also well suited to handle non-smooth objective functions