Using the Curve Fitting and Parameter Estimation Study Steps
In Part 7 of this course on optimization, we start with the topic of weighted objective functions, using the microchannel dispersion tutorial model from the previous part. We then review the settings for the acoustic demultiplexer model, which is a shape optimization example using a minimax objective.
Next, we demonstrate how to extract material parameters from experimental data using parameter estimation, and then demonstrate using the General Parameter Estimation app. Following this, you will see how to perform parameter estimation using a modified version of the elbow bracket tutorial model. We show how to approach solving the model, first using the Curve Fitting study step and then the Parameter Estimation study step. In doing so, we also show how to import external data into the software and how to take that data and plot it against your computed/simulated data and results in order to see how closely they match. We then discuss the use cases for the Curve Fitting study step and Parameter Estimation study step and what types of problems each are best suited to solve.
Further Learning
In COMSOL Multiphysics, under the Application Libraries, there are several tutorial models we recommend for additional self-guided practice in performing parameter estimation. This includes:
- Parametric Optimization of an Electric Motor
- Parametric Optimization of a Tesla Microvalve
- Shape Optimization of an Acoustic Demultiplexer
- Shape Optimization of a Tesla Microvalve
- Finding Arrhenius Parameters Using Parameter Estimation
- Impedance Tube Parameter Estimation with Data Generation
Submit feedback about this page or contact support here.