Decentralized diagnostics based on a distributed micro-genetic algorithm for transducer networks monitoring large experimental systems
| Autor: | M. Pezzetti, Pasquale Cimmino, Carlo Manna, Pasquale Arpaia, Domenico Maisto, M. Girone, G. La Commara |
|---|---|
| Přispěvatelé: | Arpaia, Pasquale, P., Cimmino, M., Girone, G., Commara, D., Maisto, C., Manna, M., Pezzetti |
| Rok vydání: | 2014 |
| Předmět: |
education.field_of_study
Large Hadron Collider Computer science Reliability (computer networking) Distributed computing Population Time efficiency Industrial research Set (abstract data type) Transducer Genetic algorithm Decentralized diagnostics distributed micro-genetic algorithm transducer networks monitoring education Instrumentation |
| Zdroj: | Review of scientific instruments 85 (2014). doi:10.1063/1.4894210 info:cnr-pdr/source/autori:Arpaia, Pasquale; Cimmino, Pasquale; Girone, Mario; Commara, Giuseppe La; Maisto, Domenico; Manna, Carlo; Pezzetti, Marco/titolo:Decentralized diagnostics based on a distributed micro-genetic algorithm for transducer networks monitoring large experimental systems/doi:10.1063%2F1.4894210/rivista:Review of scientific instruments/anno:2014/pagina_da:/pagina_a:/intervallo_pagine:/volume:85 |
| ISSN: | 1089-7623 |
| DOI: | 10.1063/1.4894210 |
| Popis: | Evolutionary approach to centralized multiple-faults diagnostics is extended to distributed transducer networks monitoring large experimental systems. Given a set of anomalies detected by the transducers, each instance of the multiple-fault problem is formulated as several parallel communicating sub-tasks running on different transducers, and thus solved one-by-one on spatially separated parallel processes. A micro-genetic algorithm merges evaluation time efficiency, arising from a small-size population distributed on parallel-synchronized processors, with the effectiveness of centralized evolutionary techniques due to optimal mix of exploitation and exploration. In this way, holistic view and effectiveness advantages of evolutionary global diagnostics are combined with reliability and efficiency benefits of distributed parallel architectures. The proposed approach was validated both (i) by simulation at CERN, on a case study of a cold box for enhancing the cryogeny diagnostics of the Large Hadron Collider, and (ii) by experiments, under the framework of the industrial research project MONDIEVOB (Building Remote Monitoring and Evolutionary Diagnostics), co-funded by EU and the company Del Bo srl, Napoli, Italy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|