Dynamic Adaption to Permanent Memory Faults in Industrial Control Systems
| Autor: | Jürgen Dobaj, Michael Krisper, Johannes Iber, Christian Kreiner |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Computer science
020206 networking & telecommunications 02 engineering and technology Industrial control system Fault (power engineering) Reliability engineering Memory cell Component (UML) Control system 0202 electrical engineering electronic engineering information engineering General Earth and Planetary Sciences 020201 artificial intelligence & image processing Software system Resilience (network) Dram General Environmental Science Dram memory |
| Zdroj: | ANT/SEIT |
| ISSN: | 1877-0509 |
| DOI: | 10.1016/j.procs.2018.04.058 |
| Popis: | Industrial control systems are making increased use of commercial off-the-shelf hardware components. One such component is memory based on DRAM technology. As pointed out by others, DRAM memory can experience permanent hardware errors, e.g. a memory cell can be permanently stuck-at zero or one. In the worst case, such a fault may have serious safety-related consequences. In this work, we present the application of a self-adaptive software system named Scari that detects erroneous datapoints, analyzes them concerning permanent stuck-at faults, and adapts to them by masking defect memory areas. Crucial for this to work is a hot-standby device that takes over the control loop during the detection and adaption phases. The goal of the mechanism presented here is automatic self-repair of a faulty control device to increase its service life and to strengthen overall resilience. The industrial setting of the presented approach is that of control devices for hydropower plant units. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|