Počet záznamů: 1
Multi-objective, rule and preference-based placement of quality sensors in water supply networks
- 1.
SYSNO ASEP 0561928 Druh ASEP C - Konferenční příspěvek (mezinárodní konf.) Zařazení RIV D - Článek ve sborníku Název Multi-objective, rule and preference-based placement of quality sensors in water supply networks Tvůrce(i) Brentan, B. (BR)
Carpitella, Silvia (UTIA-B) ORCID
Izquierdo, J. (ES)
Montalvo, I. (DE)Celkový počet autorů 4 Zdroj.dok. IFAC-PapersOnLine. Volume 55, Issue 6 - 11th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2022. - Amsterdam : Elsevier, 2022 - ISSN 2405-8963 Rozsah stran s. 482-489 Poč.str. 8 s. Forma vydání Tištěná - P Akce IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2022 /11./ Datum konání 08.06.2022 - 10.06.2022 Místo konání Pafos Země CY - Kypr Typ akce WRD Jazyk dok. eng - angličtina Země vyd. NL - Nizozemsko Klíč. slova water quality ; multi-objective optimization ; contaminant intrusion ; active diagnosis ; sensor placement ; condition monitoring ; rule-based agent ; safety-critical networked system Vědní obor RIV JC - Počítačový hardware a software Obor OECD Computer hardware and architecture Institucionální podpora UTIA-B - RVO:67985556 UT WOS 000858758500037 EID SCOPUS 85137031027 DOI 10.1016/j.ifacol.2022.07.175 Anotace To detect contaminant intrusion and, in general, to assess quality problems in their water distribution systems, water utilities need quality sensors that continuously measure, directly from the network, conductivity, PH, concentration of different substances, and other related parameters. Due to the nature of the objectives involved, the decision about where to place sensors in the network and the amount of them to be installed is a very challenging problem. In this investigation, we present a multi-objective approach to cast light on those decisions. Instead of a crisp solution, the multi-objective approach will provide a wide spectrum of solutions representing the best trade-off among all the decision criteria of the problem. This approach aims to integrate the practical experience of engineers into the decision-making process since, eventually, the solution will be selected among the Pareto front of solutions using the engineers' experience and the specific characteristics of their utility. To this end, the used algorithm adds agents based on both technical and user-preference rules on top of evolutionary search techniques to explore the decision space. The algorithm runs as a part of the Agent Swarm Optimization framework, a consolidated multi-objective software. Another novelty of this contribution is computational: the evaluation of the objective functions is executed directly in the MS SQL server and simulation data is never required to be loaded in their entirety. Without this important implementation detail, the solution for “large” water network models would not be affordable with the hardware typically used in desktop computers. To illustrate the solution process, a use case focused on a mid-size water supply network is addressed. Pracoviště Ústav teorie informace a automatizace Kontakt Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Rok sběru 2023
Počet záznamů: 1