Software for Delay Systems (SSD)

SSD is a MATLAB package providing the following features and functions for retarded time delay systems:

• Time and frequency domain visualizations: step, bode, bodemag, sigma, nyquist.
• Balanced Reduction: gram, balreal, balred (standard and frequency-limited).
• H2 norm computation: h2norm.

This repo includes three executable notebooks

• introduction.mlx overviews the features of SSD and
• benchmarks_model_reduction.mlx and benchmarks_h2norm.mlx collects benchmark problems for model reduction and H2 norm computation of delay systems and uses SSD if applicable.

Requirements

SSD requires MATLAB R2020b or newer.

Installation

Add src directory to the MATLAB's path.

Getting started

Check out the executable notebook introduction.mlx for the overview of the SSD functionalities. A snapshot of the executable notebook can be seen below

Benchmarks

Model reduction

The executable notebook benchmarks_model_reduction.mlx collects five benchmark problems.

A snapshot of the executable notebook can be seen below.

H2 norm computation

The executable notebook benchmarks_h2norm.mlx collects six benchmark problems.

A snapshot of the executable notebook can be seen below.

License

MIT License.

Citations

If you use SSD or benefit from any benchmarks code, please cite our paper:

@article{gumussoy:22:ifac_tds,
  title={SSD - Software for Systems with Delays: Reproducable Examples and Benchmarks on Model Reduction and H2 Norm Computation},
  author={Suat Gumussoy},
  journal={accepted to Joint IFAC Conference: SSSC-TDS-LPVS, Montreal, Canada, 2022},
  volume={},
  number={},
  pages={},
  year={},
  publisher={}
}