Naslov (srp)

Detekcija otkaza kotrljajnih ležajeva primenom naprednih vremensko-frekvencijskih metoda analize signala vibracija

Autor

Tsiafis, Hristos J., 1983-

Doprinosi

Todorović, Petar, 1969-
Mačužić, Ivan, 1971-
Xenos, Thomas, 1955-
Mitrović, Slobodan, 1967-
Milisavljević, Stevan.
Živić, Fatima, 1970-

Opis (srp)

Rezime: Od samih početaka industrijske revolucije pojavila se potreba za što efikasnijom i pre svega jeftinijom proizvodnjom proizvoda koja će pri tom imati željeni visok nivo kvaliteta. U tu svrhu, kreiraju se sve brže i snažnije mašine i mehanizmi, koji postaju sve kompleksniji, a samim tim izloženiji višim nivoima opterećenja. Kao rezultat toga mašine su sve više izložene kompleksnijim tipovima oštećenja i/ili otkazima koji direktno utiču na njihovu pouzdanost, raspoloživost i bezbednost pri korišćenju. Takvu opremu srećemo u finansijski i tehnički posebno kritičnim oblastima kao što su: procesi obrade, transportni sistemi, električna i elektronska oprema, elektroenergetski sistemi, a u novije vreme su to sistemi za proizvodnju obnovljive energije. Rotacione mašine spadaju u klasu najčešće korišćenih tehničkih sistema, za koje se često zahteva kompletna i precizna dokumentacija o vibracionim karakteristikama, uključujući merenja neophodna kako bi se izvršila analiza vibracija vratila, kućišta i kotrljajnih ležajeva. Kotrljajni ležajevi, reduktori i rotori su ključne i neizostavne komponente rotacionih mašina. Samim tim, stanje ovih ključnih komponenata ujedno određuje i stanje same rotacione mašine. U disertaciji se proučava primena postprocesionih metoda izdvajanja spektra iz signala vibracija, kako bi se detektovala oštećenja kotrljajnih ležajeva. Kotrljajni ležajevi su osnovne komponente rotacionih mašina. Oštećenja kotrljajnih ležajeva su odgovorna za značajan deo otkaza mašina. Samim tim otkrivanje oštećenja kotrljajnih ležajeva je važno za poboljšanje pouzdanosti i performansi mehaničkih sistema. Iako su kotrljajni ležajevi detaljno proučavani tokom prethodnih decenija što je prikazano u dostupnoj literaturi, u ovoj disertaciji je predstavljena inovativna primena vremensko–frekvencijske metode za analizu, tzv. metode kompletne ansambalske empirijske dekompozicije na modove sa adaptivnim šumom (eng. Complete Ensemble Empirical Mode Decomposition with Adaptive Noise – CEEMDAN) i Zao–Atlas–Markove raspodele (ZAMD). Pomenutim metodama se prevazilaze poznata ograničenja metode razlaganja signala na funkcije (eng. Empirical Mode Decomposition – EMD), u pogledu mešanja modova i izdvajanja frekvencija. Glavni cilj disertacije je istraživanje sposobnosti ovih metoda za otkrivanje oštećenja u početnoj fazi. Za procenu metoda, koriste se kotrljajni ležajevi sa poznatim lokalizovanim oštećenjima, u cilju dobijanja skupa podataka sa eksperimentalnog postrojenja koje simulira rotacionu mašinu. Isto tako, upotrebljavaju se i podaci iz literature kao drugi test. Dobijeni rezultati potvrđuju sposobnost metode za otkrivanje degradacije kotrljajnih ležajeva.

Opis (eng)

Abstract: Since the industrial revolution, a need for faster, better quality and especially cheaper to produce, products has emerged. So special tools, quick and powerful machines and mechanisms were created, and tend to become increasingly complex, thus subject to a corresponding complex damage and / or failures affecting the reliability, availability and safety of operation. Such equipment’s are found in particularly critical financial and technical fields such as machining processes, production, transport systems, electrical and electronic equipment and power systems (and, recently, renewable energy). Rotating machinery is one of the most common classes of machines, often requires complete and accurate documentation of vibration characteristics including measurements for shaft, housing and rolling bearings vibration analysis. Rolling bearings, gears and rotors are the common and key components in rotating machinery. The health condition of these key components represents that of the machine itself. The present dissertation introduces investigates the application of a post-processing method of extracting spectra from vibration signals in order to detect faults of rollingelement bearings. Rolling-element bearings are fundamental components of rotating machinery. Faults of rolling-elements bearings are responsible for a substantial proportion of machine failures and therefore fault detection is important for improving the mechanical system reliability and performance. Although rolling bearings have been investigated in detail in past studies, innovative applications of time-frequency analysis method, called complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and Zhao Atlas Marks Distribution (ZAMD), that overcomes known limitations concerning mode mixing and frequency separation of empirical mode decomposition are presented. The main aim of the presented dissertation is to investigate the ability of the methods to detect faults in early stage. To validate the methods, rollingelement bearings with known and localized faults are used in order to acquire datasets from an experimental rig that stimulates rotating machinery. Also datasets from literature are used as second trial. The results verify the ability of the method to detect degradations of rolling-element bearings. The present dissertation consists of six experimental studies and one industrial case study, where the first one was concerned with investigation and validation of the experimentalrig in relation to its dimensions and construction accuracy and the second one concerned the validation and selection of mounting pads in order to isolate the experimental-rig from external stimulations and the calculation of their mechanical properties. The four remaining experimental studies were concerned with investigation and validation of advanced signal processing methodologies for their ability to detect external stimulations and to detect faults in early stage in rolling- element bearings. At last an industrial case study was conducted in order to validate the method in real working environment.

Jezik

srpski

Datum

2018

Licenca

Creative Commons licenca
Ovo delo je licencirano pod uslovima licence
Creative Commons CC BY-NC-ND 2.0 AT - Creative Commons Autorstvo - Nekomercijalno - Bez prerada 2.0 Austria License.

CC BY-NC-ND 2.0 AT

http://creativecommons.org/licenses/by-nc-nd/2.0/at/

Identifikatori