Data sets from CPS for ML applications
Structural Analysis is a powerful tool in fault diagnosis that abstracts the dynamic equations modeling the system to only the links between equations and variables. Many complex algorithms are developed to work with structural models, especially to generate sets of relations leading to diagnostic tests, i.e. Analytical Redundancy Relations or parity equations, or to select the most relevant ones. How- ever, there is a lack of easily accessible large benchmarks for comparing and testing the algorithm’s computational com- plexity and memory consumption. In this paper, we present a set of structural models of water distribution networks with a wide range of sizes and a Python library for easy and configurable generation of structural models from wa- ter network description.
| Title | Authors | Year |
|---|---|---|
| Water network benchmarks for structural analysis algorithms in fault diagnosis | Sztyber et al. | 2022 |
anna.sztyber@pw.edu.pl