56 results on '"Milivojević, Nikola"'
Search Results
52. EMBANKMENT DAM STABILITY ANALYSIS USING FEM
- Author
-
Rakić, Dragan, primary, Živković, Miroslav, additional, Vulović, Snežana, additional, Divac, Dejan, additional, Slavković, Radovan, additional, and Milivojević, Nikola, additional
- Published
- 2013
- Full Text
- View/download PDF
53. DEVELOPMENT OF NEURO-FUZZY MODEL FOR DAM SEEPAGE ANALYSIS.
- Author
-
NOVAKOVIĆ, Aleksandar, RANKOVIĆ, Vesna, GRUJOVIĆ, Nenad, DIVAC, Dejan, and MILIVOJEVIĆ, Nikola
- Subjects
DAMS ,SOIL infiltration ,WATER levels ,COMPUTATIONAL intelligence ,FUZZY logic - Abstract
Modeling seepage through geological formations located near the dam site or dam bodies is a challenging task in dam engineering. In order to monitor the seepage, piezometric devices are installed on sections of the dam. The objective of this study is to develop a neuro-fuzzy model to predict the water level in piezometers of the Iron Gate 2 dam. The neuro-fuzzy model was developed using experimental data which was collected during a period of nine years. The measurements of tailwater level taken on the same day, one day before, and two days before the measurements of piezometer were input variables, and the water level in the examined piezometer was the target output in the neuro-fuzzy model. The measured data has been compared with the results of neuro-fuzzy model on the basis of correlation coefficient (r), coefficient of determination (R²), mean square error (MSE) and mean absolute error (MAE). Comparing the experimental data with the values modeled by the neuro-fuzzy system indicates that the computational intelligence models provide very accurate results. [ABSTRACT FROM AUTHOR]
- Published
- 2014
54. PRISTUPAČNA EDUKATIVNA PLATFORMA ZASNOVANA NA PROŠIRENOJ REALNOSTI.
- Author
-
Milivojević, Nikola, Stamatović, Nikola, Stojanović, Nemanja, and Krstin, Milan
- Subjects
EDUCATION ,RESEARCH ,COMMUNICATION ,LEARNING ,TEACHING - Abstract
Copyright of InfoM is the property of Belgrade University, Faculty of Organizational Science and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2011
55. Metodologija za brzu asimilaciju podataka u modelima otvorenih tokova
- Author
-
Milašinović, Miloš R., Prodanović, Dušan, Savić, Dragan, Stanić, Miloš, Zindović, Budo, and Milivojević, Nikola
- Subjects
Model-driven forecasting, Data Assimilation, 1D open channel flow modelling, Proportional-Integrative-Derivative controllers, flow hydrograph reconstruction ,Prognoze vođene modelima, Asimilacija podataka, linijski (1D) modeli otvorenih tokova, PID kontroleri, rekonstukcija hidrograma - Abstract
Upravljanje vodnim resursima (vodosnabdevanje, odbrana od poplava, hidroenergetika) zahteva prognoze dostupne količine vode kao pomoć u donošenju upravljačkih odluka. Najčešće se te prognoze zasnivaju na upotrebi različitih fizički zasnovanih modela. Tako dobijeni rezultati su često veoma nepouzdani zbog prisustva različitih tipova neodređenosti. Kod primene linijskih (1D) modela tečenja u otvorenim tokovima neki od dominantnijh izvora neodređenosti su nedovoljno dobro poznavanje graničnih uslova (hidrogrami dotoka, krive protoka) i nepouzdani početni uslovi koji su osnova od koje se započinje prognoza. Kontinulana merenja na sistemu (npr. merenja nivoa na reci) predstavljaju relativno pouzdan reprezent njegovog stanja ali pokrivaju samo jedan mali deo domena koji se razmatra. Zbog toga se pribegava postupku asimilacije podataka kojom se usaglašavaju rezultati modela i merenja. Ova usaglašavanja se sprovode za već prikupljene podatke, u vremenskom periodu neposredno pre trenutka od koga se želi izvršiti prognoza. Taj period usaglašavanja naziva se asimilacioni period. Na kraju asimilacionog perioda model bolje oslikava trenutno stanje na celom domenu i može se iskoristiti za prognozu. U ovom istraživanju prikazana je nova asimilaciona metoda za modele otvorenih tokova korišćenjem indirektnog, fizički zasnovanog pristupa za usaglašavanje nivoa pomoću korektivnih dotoka. Korektivni dotoci se računaju koristeći teoriju Proporcionalno-Integrativno-Derivativnih (PID) kontrolera iz oblasti teorije upravljanja (eng. control theory). Metoda je razvijena u cilju primene jednostavnijeg i bržeg postupka asimilacije podataka u realnim sistemima otvorenih tokova. Nova asimilaciona metoda poredi se sa standardnom metodom asimilacije podataka (Ensemble Kalman Filter - EnKF) koja se najčešće koristi u sličnim istraživanjima. Poređenjem na hipotetičkim test primerima pokazano je da se primenom nove asimilacione metode dolazi do značajnog ubrzanja proračuna bez smanjenja kvaliteta asimilacije podataka. Primena ove metode na realne sisteme zahteva rešavanje problema određivanja optimalne konfiguracije (forme, strukture) i optimalnog podešavanja parametara PID kontrolera. Na primeru Hidroenergetskog sistema Đerdap sprovedeno je detaljno ispitivanje optimalne konfiguracije na osnovu definisanih indikatora kvaliteta asimilacije podataka. Utvrđeno je da jednostavnija konfiguracija kontrolera, sastavljena od proporcionalnog i integrativnog faktora (PI kontroler) daje najbolje rezultate asimilacije. Optimalno podešavanje parametara kontrolera rešeno je višekriterijumskom optimizacijom pomoću genetskog algoritma sa nedominantnim sortiranjem (eng. Nondominated Sorting Genetic Algorithm II – NSGA-II). Utvrđeno je da se podešavanje kontrolera višekriterijumskom optimizacijom može uraditi koristeći neku od kombinacija sa 2 suprotstavljene kriterijumske funkcije. Dodatna ispitivanja pokazala su potencijal ove asimilacione metode za rekonstrukciju stvarnih hidrograma na osnovu zabeleženih nivograma, kao alata za rekonstrukciju poplavnih talasa i smanjivanje neodređenosti krivih protoka. Water resources management (water supply, flood prevention, hydropower) requires many forecasts of the water available, as a decision-support tool. These forecasts are provided using physically based numerical models (model-driven forecasting) and can produce results of unsatisfying accuracy due to numerous uncertainties. When 1D open channel flow models are used, some of the dominating uncertainty sources are unreliable boundary conditions (inflow hydrograph, rating curves) and initial conditions. Continuous observations of the system state can be considered as a reliable representation of the true state but limited to the small domain. Hence, data assimilation is used for coupling the measurements and models. The goal of data assimilation is to reduce the difference between measured and modelled data (e.g. water levels) by continuous model results update (water level update). This update is conducted for the collected data in the period before the moment used for the start of the forecast, called assimilation period (assimilation window). Water level update provides improved initial conditions at the end of this period. This research presents a novel data assimilation method based on the indirect, physically based approach by adding/subtracting correction flows from the model. These correction flows are calculated using the control theory, specifically Proportional-Integrative-Derivative (PID) controller theory. The aim of this method is to propose a simplified and faster assimilation procedure without sacrificing the accuracy. Proposed data assimilation method is compared to the widely used Ensemble Kalman Filter (EnKF). Benchmark test on hypothetical test cases shows substantial speed up when PID controllers are used as data assimilation method, without significant sacrifice of the accuracy. Further application of this method on real world problems required thorough investigation on optimal controllers’ configuration and optimal tuning of the controllers. Phase analysis is conducted to determine optimal controllers’ configuration, based on the data assimilation quality indicators. Analysis shows that proportional-integrative configuration (PI controller) of the controllers should be used only, without derivative gain. Research shows the potential of using multicriteria optimization for optimal tuning of the controllers’ parameters, using Nondominated Sorting Genetic Algorithm (NSGA-II) for the minimization of the criterias. Analysis also shows that multicriteria used for optimal tuning of the controllers should not consider more than two conflicting criterias. Further investigation shows the potential of using the novel data assimilation method for true flow hydrograph reconstruction. This algorithm could be used as a tool for inverse flood routing and uncertainty reduction of the existing rating curves.
- Published
- 2021
56. Tehnike računarske inteligencije u modeliranju i identifikaciji indikatora ponašanja brane
- Author
-
Aleksandar B. Novaković, Ranković, Vesna, Divac, Dejan, Grujović, Nenad, Živković, Miroslav, and Milivojević, Nikola
- Subjects
indikatori ponašanja ,identifikacija ,neuro-fazi sistem ,neuronske mreže - Abstract
Indikatori ponašanja brane su relevantne veličine, čijim se praćenjem utvrđuje da li je stvarno stanje brane u eksploataciji u saglasnosti sa onim što je predviđeno i očekivano u fazi projektovanja. Veličine koje se prate treba da se kreću u nekom unapred definisanom opsegu koji garantuje stanje stabilnosti brane. U ovoj disertaciji su predloženi različiti pristupi modeliranja i parametarske identifikacije indikatora ponašanja brane, poput horizontalnih pomeranja i nivoa vode u pijezometrima, tehnikama računarske inteligencije. Prvi pristup je da se linearno preslikavanje uzročnih veličina u indikatore ponašanja, koje se koristi kod višestruke linearne regresije, zameni nelinearnim. Drugi pristup, predložen u ovom radu, zasniva se na primeni postupka parametarske identifikacije nelinearnih sistema. Horizontalna pomeranja i nivoi vode u pijezometrima su nelinearne, složene funkcije uzročnih veličina, pa je za njihovo modeliranje korišćena NARX (Nonlinear Auto Regresive eXogenous- nelinearni auto-regresioni model sa spoljašnjim ulazom) struktura, kojom je opisana široka klasa nelinearnih dinamičkih procesa. Predloženi pristupi formiranja modela primenjeni su za modeliranje i parametarsku identifikaciju horizontalnih pomeranja tačaka brane Bočac, kao i nivoa vode u pijezometrima brana Đerdap II i Prvonek. Nelinearni modeli zasnovani na tehnikama računarske inteligencije implementirani su korišćenjem programskog jezika Java i programskog paketa Matlab. Tehnike računarske inteligencije korišćene u ovom radu su višeslojni perceptron, RBF (RBF - Radial Basis Function – radijalna osnovna funkcija) neuronska mreža i ANFIS (ANFIS - Adaptive-Network-Based Fuzzy Inference System - fazi sistem za zaključivanje zasnovan na adaptivnoj mreži). Nedostajući podaci u skupu merenja mogu biti uzrok problema u okviru procesa učenja i loših performansi dobijenih modela. U cilju nadomeštanja nedostajućih podataka korišćene su tehnike iz domena matematičke statistike. Prisustvo autlajera u mernim podacima ima veliki uticaj na predviđanja podataka koji nedostaju, pa je njihovo prisustvo posebno analizirano. Takođe je analiziran i problem optimizacije ulazno-izlaznih modela, koji podrazumeva određivanje broja prediktora i dimenzije regresionog vektora, kao i broja parametara neuronskih mreža i neuro-fazi sistema. Performanse modela, formiranih na osnovu predloženog koncepta, poređeni su sa rezultatima dobijenim drugim metodama modeliranja istih indikatora ponašanja prikazanim u relevantoj literaturi objavljenoj u poslednjih nekoliko godina. Na osnovu rezultata zaključeno je da je moguće kreirati i obučiti modele zasnovane na tehnikama računarske inteligencije koji će sa velikom preciznošću predviđati bitne indikatore ponašanja brane. The dam behavior indicators are relevant factors whose monitoring indicates whether the actual operational state of the dam is in accordance with what is expected and anticipated in the design phase. Such indicators should move in a predefined range, in order to guarantee stability of the dam. This dissertation proposes different approaches to modeling and parametric identification of the dam behavior indicators, such as radial displacements or piezometric water levels, using the techniques of artificial intelligence. The first approach is to replace linear mapping of causal variables into behavior indicators, which is used in multiple linear regression, with nonlinear. The second approach proposed in this paper is based on applying the method of parametric nonlinear system identification. Radial displacements and piezometric water levels are nonlinear, complex functions of causal variables, so for their modeling NARX (Nonlinear Auto Regresive eXogenous), which is employed to describe a wide class of nonlinear dynamic systems, is used. These proposed approaches are used for modeling and parametric identification of radial displacements of dam Bočac, and piezometric water levels of dams Iron Gate II and Prvonek. Nonlinear models based on artificial intelligence techniques have been implemented using the Java programming language and MATLAB. Artificial intelligence techniques used in this work are the multilayer perceptron, RBF (Radial Basis Function) neural network and ANFIS (Adaptive-Network-Based Fuzzy Inference System). The presence of missing data in a set of measurements may be causing problems in the learning process and the poor performance of the obtained models. In order to predict the missing data, the techniques of mathematical statistics have been used. Outliers present in a set of measurements have a big effect on the prediction of missing data, and their presence is specifically analyzed. The problem of optimizing the inputoutput model, which involves determining the number of predictors and dimensions of the regression vector, and the number of parameters of neural networks and neuro-fuzzy systems, is also analyzed. The performance of the models, formed on the basis of the proposed concept, are compared with those obtained by other methods of modeling the same behavioral indicators presented in relevant accompanying literature published in the last few years. Based on the results, it was concluded that it is possible to create and train models based on computational intelligence techniques to predict with great accuracy the essential dam behavior indicators.
- Published
- 2014
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.