Mjerna metodologija temeljena na modelu s primjenom na ispitivanje uzemljivačkih sustava

doctoral thesis

• Title (english): Model-Based Measurement with Application to Grounding System Testing
• Author: Hrvoje Džapo
• Mentor: Stanko Tonković (mentor)
• Committee member: Željko Štih (predsjednik povjerenstva)
• Committee member: Stanko Tonković (član povjerenstva)
• Committee member: Srđan Žutobradić (član povjerenstva)
• Committee member: Ljubo Marangunić (član povjerenstva)
• Committee member: Sead Berberović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronic Systems and Information Processing); Zagreb
• Defense date and country: 2007-05-30, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics; Probability Theory and Statistics
• Universal decimal classification (UDC): 51 - Mathematics ; 621.3 - Electrical engineering

Abstract

Postojećim postupcima numeričkog proračuna uzemljivača izračunata je referentna raspodjela potencijala na površini tla. Provedena je analiza kvalitete rekonstrukcije raspodjele potencijala primjenom standardnih interpolacijskih postupaka, uz različite geometrijske rasporede uzoraka i variranje utjecaja smetnji slučajnog karaktera. Diskutirani su uočeni nedostaci u postojećoj praksi rekonstrukcije i definirani zahtjevi na unaprijeđeni algoritam interpolacije nedostajućih vrijednosti potencijala, u skladu s potrebama mjernog postupka ispitivanja stanja sigurnosti uzemljivačkog sustava. Teorijski je razrađen i programski implementiran novi, na fizikalnom modelu utemeljen pristup rekonstrukciji raspodjele potencijala na površini tla iz diskretnog skupa izmjerenih vrijednosti, zasnovan na stohastičkom pristupu rješavanju inverznog problema u statičkom strujnom polju uzemljivača. Vrednovanje novorazvijene metodologije provedeno je analizom kvalitete rekonstrukcije teorijski predviđene raspodjele potencijala, pri čemu su usporedbom s rezultatima za klasične interpolacijske postupke dokazane prednosti novog pristupa. Razrađena je nova metodologija dizajna optimalnog eksperimenta za mjerenje raspodjele potencijala na površini tla, zasnovana na Bayesovom stohastičkom pristupu i formalizmu mjera iz teorije informacija. Valjanost novorazvijene metodologije planiranja učinkovitog eksperimenta potvrđena je kroz više oglednih simulacija, a također su prikazane mogućnosti interaktivne optimizacije eksperimenta. Predložena su jednostavna heuristička pravila za planiranje optimalnog rasporeda mjernih točaka, teorijski utemeljena na razrađenom pristupu eksperimentalnom dizajnu.

Abstract (english)

Reference earth surface potential distribution was calculated using well-known numerical models. The potential distribution reconstruction quality analysis for some of the most commonly used scattered data interpolation techniques was performed, depending on different sampling geometries and variable measurement noise levels. Perceived shortcomings of standard interpolation methods for earth surface potential distribution reconstruction were discussed and essential requirements for improvement of the reconstruction methodology were formulated, in accordance with the needs of the grounding grid safety assessment measuring procedure. The novel model-based approach to the potential distribution reconstruction from a discrete set of noisy measuring samples by solving stochastic inverse problem in Bayesian paradigm was theoretically elaborated and implemented in software. The new methodology was evaluated through the simulations and the potential distribution quality analysis. The results justified advantages of the new approach, which exhibits superiority over standard scattered data interpolation techniques. The novel optimal experimental design methodology for sampling earth surface potential distribution was elaborated, based on the Bayesian stochastic approach and the information theory formalism. The validity of the new experimental design approach was verified by simulations and some possibilites for interactive experimental design were presented. Simple theoretically justified heuristic rules for optimizing sampling geometries in the earth surface potential distribution survey were presented.

• Keywords: sustav uzemljenja; napon koraka i dodira; mjerenje raspodjele potencijala na površini tla; senzorske mreže; interpolacijske metode; inverzni problem; Bayesov pristup rješavanju inverznog problema; Monte Carlo Markovljevi lanci; dizajn optimalnog eksperimenta; Fisherova informacija
• Keywords (english): grounding system; touch and step voltage; earth surface potential distribution measurement; sensor arrays; interpolation methods; inverse problem; Bayesian approach to the inverse problem solution; Monte Carlo Markov chains; optimal experimental design; Fisher information
• Language: croatian
• URN:NBN: urn:nbn:hr:168:232183
• Study programme: Title: Doctoral study programme "Electrical Engineering and Computing"; Study programme type: university; Study level: postgraduate; Academic / professional title: doktor/doktorica znanosti, po-dručje tehničkih znanosti (doktor/doktorica znanosti, po-dručje tehničkih znanosti)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-04-06 07:37:07