Simulink Design Optimization 1.1

Introduction

Simulink Design Optimization™ lets you improve designs by estimating and tuning model parameters using numerical optimization. You can increase model accuracy by using test data to calibrate physical parameters, such as mass or resistance. You can then improve system performance, reduce system cost, and meet other objectives by automatically tuning design parameters in your Simulink® model. For example, you can optimize controller gains to meet rise-time and overshoot constraints, or jointly optimize physical and algorithmic parameters to maximize overall system performance.

Key Features

• Estimation of physical parameters from test data
• Preprocessing of test data, including data selection, offset removal, detrending, noise filtering, and missing data reconstruction
• Optimization of time-domain responses of nonlinear Simulink models
• Optimization of time- and frequency-domain responses of linear control systems (with Control System Toolbox™)
• Ability to graphically specify response requirements and then visually monitor the optimization progress
• Ability to factor in parameter variation or uncertainty for robust design optimization

Using Simulink Design Optimization with measured data to calibrate parameters for the blocks highlighted in red. The lower left and lower right images show measured versus simulated data before and after the calibration, respectively. Click on image to see enlarged view.