TY  - JOUR
AU  - Kostić, Srđan
AU  - Vasović, Dejan
PY  - 2015
UR  - https://raf.arh.bg.ac.rs/handle/123456789/204
AB  - In the present paper, we propose a prediction model for concrete compressive strength using artificial neural networks. In experimental part of the research, 75 concrete samples with various w/c ratios were exposed to freezing and thawing, after which their compressive strength was determined at different age, viz. 7, 20 and 32 days. In computational phase of the research, different prediction models for concrete compressive strength were developed using artificial neural networks with w/c ratio, age and number of freeze/thaw cycles as three input nodes. We examined three-layer feed-forward back-propagation neural networks with 2, 6 and 9 hidden nodes using four different learning algorithms. The most accurate prediction models, with the highest coefficient of determination (R (2) > 0.87), and with all of the predicted data falling within the 95 % prediction interval, were obtained with six hidden nodes using Levenberg-Marquardt, scaled conjugate gradient and one-step secant algorithms, and with nine hidden nodes using Broyden-Fletcher-Goldfarb-Shannon algorithm. Further analysis showed that relative error between the predicted and experimental data increases up to acceptable a parts per thousand 15 %, which confirms that proposed ANN models are robust to the consistency of training and validation output data. Accuracy of the proposed models was further verified by low values of standard statistical errors. In the final phase of the research, individual effect of each input parameter was examined using the global sensitivity analysis, whose results indicated that w/c ratio has the strongest impact on concrete compressive strength.
PB  - Springer-Verlag London Ltd
T2  - Neural computing & applications
T1  - Prediction model for compressive strength of basic concrete mixture using artificial neural networks
VL  - 26
IS  - 5
SP  - 1005
EP  - 1024
DO  - 10.1007/s00521-014-1763-1
ER  - 

@article{
author = "Kostić, Srđan and Vasović, Dejan",
year = "2015",
abstract = "In the present paper, we propose a prediction model for concrete compressive strength using artificial neural networks. In experimental part of the research, 75 concrete samples with various w/c ratios were exposed to freezing and thawing, after which their compressive strength was determined at different age, viz. 7, 20 and 32 days. In computational phase of the research, different prediction models for concrete compressive strength were developed using artificial neural networks with w/c ratio, age and number of freeze/thaw cycles as three input nodes. We examined three-layer feed-forward back-propagation neural networks with 2, 6 and 9 hidden nodes using four different learning algorithms. The most accurate prediction models, with the highest coefficient of determination (R (2) > 0.87), and with all of the predicted data falling within the 95 % prediction interval, were obtained with six hidden nodes using Levenberg-Marquardt, scaled conjugate gradient and one-step secant algorithms, and with nine hidden nodes using Broyden-Fletcher-Goldfarb-Shannon algorithm. Further analysis showed that relative error between the predicted and experimental data increases up to acceptable a parts per thousand 15 %, which confirms that proposed ANN models are robust to the consistency of training and validation output data. Accuracy of the proposed models was further verified by low values of standard statistical errors. In the final phase of the research, individual effect of each input parameter was examined using the global sensitivity analysis, whose results indicated that w/c ratio has the strongest impact on concrete compressive strength.",
publisher = "Springer-Verlag London Ltd",
journal = "Neural computing & applications",
title = "Prediction model for compressive strength of basic concrete mixture using artificial neural networks",
volume = "26",
number = "5",
pages = "1005-1024",
doi = "10.1007/s00521-014-1763-1"
}

Kostić, S.,& Vasović, D.. (2015). Prediction model for compressive strength of basic concrete mixture using artificial neural networks. in Neural computing & applications
Springer-Verlag London Ltd., 26(5), 1005-1024.
https://doi.org/10.1007/s00521-014-1763-1

Kostić S, Vasović D. Prediction model for compressive strength of basic concrete mixture using artificial neural networks. in Neural computing & applications. 2015;26(5):1005-1024.
doi:10.1007/s00521-014-1763-1 .

Kostić, Srđan, Vasović, Dejan, "Prediction model for compressive strength of basic concrete mixture using artificial neural networks" in Neural computing & applications, 26, no. 5 (2015):1005-1024,
https://doi.org/10.1007/s00521-014-1763-1 . .

TY  - JOUR
AU  - Kostić, Srđan
AU  - Vasović, Dejan
PY  - 2015
UR  - https://raf.arh.bg.ac.rs/handle/123456789/184
AB  - In present paper, concrete compressive strength is evaluated using back propagation feed-forward artificial neural network. Training of neural network is performed using Levenberg-Marquardt learning algorithm for four architectures of artificial neural networks, one, three, eight and twelve nodes in a hidden layer in order to avoid the occurrence of overfitting. Training, validation and testing of neural network is conducted for 75 concrete samples with distinct w/c ratio and amount of superplasticizer of melamine type. These specimens were exposed to different number of freeze/thaw cycles and their compressive strength was determined after 7, 20 and 32 days. The obtained results indicate that neural network with one hidden layer and twelve hidden nodes gives reasonable prediction accuracy in comparison to experimental results (R=0.965, MSE=0.005). These results of the performed analysis are further confirmed by calculating the standard statistical errors: the chosen architecture of neural network shows the smallest value of mean absolute percentage error (MAPE=, variance absolute relative error (VARE) and median absolute error (MEDAE), and the highest value of variance accounted for (VAF).
AB  - U radu se daje procena čvrstoće betona pri pritisku, primenom veštačkih neuronskih mreža s prostiranjem signala unapred i propagacijom greške unazad. Obučavanje mreže sprovodi se korišćenjem Levenberg-Markart algoritma obučavanja za četiri različite arhitekture neuronskih mreža, s jednom jedinicom, tri jedinice, te osam i dvanaest jedinica u skrivenom sloju, radi odbacivanja efekta ,,pretreniranja'. Treniranje, validacija i testiranje neuronskih mreža izvodi se na osnovu rezultata eksperimentalnog ispitivanja čvrstoće pri pritisku na 75 uzoraka betona, s različitim vodocementnim faktorom i količinom superplastifikatora tipa melamina. Ispitivani uzorci betona izlagani su različitim ciklusima zamrzavanja/ otkravljivanja, a njihova čvrstoća pri pritisku određivana je nakon 7, 20 i 32 dana. Dobijeni rezultati ukazuju na to da neuronska mreža s dvanaest jedinica u skrivenom sloju daje ocenu čvrstoće zadovoljavajuće tačnosti u poređenju sa eksperimentalno dobijenim podacima (R≈0,97, MSE=0,005). Rezultati izvedene analize dodatno su potvrđeni sračunavanjem vrednosti standardnih statističkih grešaka: najmanjom vrednošću srednje apsolutne greške (MAPE), varijanse relativne vrednosti apsolutne greške (VARE) i medijane (MEDAE), kao i najvećom vrednošću sračunate varijanse (VAF) za izabranu arhitekturu neuronske mreže.
PB  - Društvo za ispitivanje i istraživanje materijala i konstrukcija Srbije, Beograd
T2  - Građevinski materijali i konstrukcije
T1  - Estimation of concrete compressive strength using artificial neural network
T1  - Procena čvrstoće betona pri pritisku, korišćenjem veštačkih neuronskih mreža
VL  - 58
IS  - 1
SP  - 3
EP  - 16
DO  - 10.5937/grmk1501003K
ER  - 

@article{
author = "Kostić, Srđan and Vasović, Dejan",
year = "2015",
abstract = "In present paper, concrete compressive strength is evaluated using back propagation feed-forward artificial neural network. Training of neural network is performed using Levenberg-Marquardt learning algorithm for four architectures of artificial neural networks, one, three, eight and twelve nodes in a hidden layer in order to avoid the occurrence of overfitting. Training, validation and testing of neural network is conducted for 75 concrete samples with distinct w/c ratio and amount of superplasticizer of melamine type. These specimens were exposed to different number of freeze/thaw cycles and their compressive strength was determined after 7, 20 and 32 days. The obtained results indicate that neural network with one hidden layer and twelve hidden nodes gives reasonable prediction accuracy in comparison to experimental results (R=0.965, MSE=0.005). These results of the performed analysis are further confirmed by calculating the standard statistical errors: the chosen architecture of neural network shows the smallest value of mean absolute percentage error (MAPE=, variance absolute relative error (VARE) and median absolute error (MEDAE), and the highest value of variance accounted for (VAF)., U radu se daje procena čvrstoće betona pri pritisku, primenom veštačkih neuronskih mreža s prostiranjem signala unapred i propagacijom greške unazad. Obučavanje mreže sprovodi se korišćenjem Levenberg-Markart algoritma obučavanja za četiri različite arhitekture neuronskih mreža, s jednom jedinicom, tri jedinice, te osam i dvanaest jedinica u skrivenom sloju, radi odbacivanja efekta ,,pretreniranja'. Treniranje, validacija i testiranje neuronskih mreža izvodi se na osnovu rezultata eksperimentalnog ispitivanja čvrstoće pri pritisku na 75 uzoraka betona, s različitim vodocementnim faktorom i količinom superplastifikatora tipa melamina. Ispitivani uzorci betona izlagani su različitim ciklusima zamrzavanja/ otkravljivanja, a njihova čvrstoća pri pritisku određivana je nakon 7, 20 i 32 dana. Dobijeni rezultati ukazuju na to da neuronska mreža s dvanaest jedinica u skrivenom sloju daje ocenu čvrstoće zadovoljavajuće tačnosti u poređenju sa eksperimentalno dobijenim podacima (R≈0,97, MSE=0,005). Rezultati izvedene analize dodatno su potvrđeni sračunavanjem vrednosti standardnih statističkih grešaka: najmanjom vrednošću srednje apsolutne greške (MAPE), varijanse relativne vrednosti apsolutne greške (VARE) i medijane (MEDAE), kao i najvećom vrednošću sračunate varijanse (VAF) za izabranu arhitekturu neuronske mreže.",
publisher = "Društvo za ispitivanje i istraživanje materijala i konstrukcija Srbije, Beograd",
journal = "Građevinski materijali i konstrukcije",
title = "Estimation of concrete compressive strength using artificial neural network, Procena čvrstoće betona pri pritisku, korišćenjem veštačkih neuronskih mreža",
volume = "58",
number = "1",
pages = "3-16",
doi = "10.5937/grmk1501003K"
}

Kostić, S.,& Vasović, D.. (2015). Estimation of concrete compressive strength using artificial neural network. in Građevinski materijali i konstrukcije
Društvo za ispitivanje i istraživanje materijala i konstrukcija Srbije, Beograd., 58(1), 3-16.
https://doi.org/10.5937/grmk1501003K

Kostić S, Vasović D. Estimation of concrete compressive strength using artificial neural network. in Građevinski materijali i konstrukcije. 2015;58(1):3-16.
doi:10.5937/grmk1501003K .

Kostić, Srđan, Vasović, Dejan, "Estimation of concrete compressive strength using artificial neural network" in Građevinski materijali i konstrukcije, 58, no. 1 (2015):3-16,
https://doi.org/10.5937/grmk1501003K . .

TY  - JOUR
AU  - Kostić, Srđan
AU  - Vasović, Dejan
AU  - Okrajnov-Bajić, Ruža
PY  - 2014
UR  - https://raf.arh.bg.ac.rs/handle/123456789/169
AB  - A new type of calcium-nitrate and urea-based chemical admixture is proposed, in order to maintain the compressive strength of fresh concrete exposed to very low temperatures (below to - 250C), including a sudden transition to positive temperatures at an early age. The applied admixture has no negative effect on compressive strength of specimens cured in water at 200C. When it is cured under three different frost regimes, concrete specimens with admixture show over three times higher compressive strength, in comparison to specimens without admixture. The implications of such improved concrete composition are discussed in reference to the tunnel lining design.
AB  - U radu se predlaže novi tip hemijskog dodatka betonu na bazi kalcijum-nitrata i uree, sa ciljem održavanja čvrstoće na pritisak svežeg betona izloženog vrlo niskim temperaturama (do -250C), uključujući i uticaj iznenadnih velikih temperaturnih amplituda (prelaze od negativnih ka pozitivnim temperaturama). Primenjeni dodatak ne utiče nepovoljno na čvrstoću na pritisak uzoraka betona negovanih u izotermalnim uslovima na temperaturi od 200C. Kada se svež beton izloži uticaju različitih režima mraza, uzorci sa dodatkom pokazuju gotovo tri puta veću pritisnu čvrstoću u poređenju sa uzorcima bez dodatka. Primena betona sa predloženim aditivom u praksi razmatra se u odnosu na postojanost i stabilnost tunelske obloge sa prskanim betonom.
PB  - Institut za rudarstvo i metalurgiju, Bor
T2  - Mining and Metallurgy Engineering Bor
T1  - New fresh concrete chemical admixture for tunnel lining design in the extreme winter conditions
T1  - Novi hemijski dodatak svežem betonu za izvođenje tunelske obloge u ekstremnim zimskim uslovima
IS  - 2
SP  - 13
EP  - 32
DO  - 10.5937/mmeb1402013k
ER  - 

@article{
author = "Kostić, Srđan and Vasović, Dejan and Okrajnov-Bajić, Ruža",
year = "2014",
abstract = "A new type of calcium-nitrate and urea-based chemical admixture is proposed, in order to maintain the compressive strength of fresh concrete exposed to very low temperatures (below to - 250C), including a sudden transition to positive temperatures at an early age. The applied admixture has no negative effect on compressive strength of specimens cured in water at 200C. When it is cured under three different frost regimes, concrete specimens with admixture show over three times higher compressive strength, in comparison to specimens without admixture. The implications of such improved concrete composition are discussed in reference to the tunnel lining design., U radu se predlaže novi tip hemijskog dodatka betonu na bazi kalcijum-nitrata i uree, sa ciljem održavanja čvrstoće na pritisak svežeg betona izloženog vrlo niskim temperaturama (do -250C), uključujući i uticaj iznenadnih velikih temperaturnih amplituda (prelaze od negativnih ka pozitivnim temperaturama). Primenjeni dodatak ne utiče nepovoljno na čvrstoću na pritisak uzoraka betona negovanih u izotermalnim uslovima na temperaturi od 200C. Kada se svež beton izloži uticaju različitih režima mraza, uzorci sa dodatkom pokazuju gotovo tri puta veću pritisnu čvrstoću u poređenju sa uzorcima bez dodatka. Primena betona sa predloženim aditivom u praksi razmatra se u odnosu na postojanost i stabilnost tunelske obloge sa prskanim betonom.",
publisher = "Institut za rudarstvo i metalurgiju, Bor",
journal = "Mining and Metallurgy Engineering Bor",
title = "New fresh concrete chemical admixture for tunnel lining design in the extreme winter conditions, Novi hemijski dodatak svežem betonu za izvođenje tunelske obloge u ekstremnim zimskim uslovima",
number = "2",
pages = "13-32",
doi = "10.5937/mmeb1402013k"
}

Kostić, S., Vasović, D.,& Okrajnov-Bajić, R.. (2014). New fresh concrete chemical admixture for tunnel lining design in the extreme winter conditions. in Mining and Metallurgy Engineering Bor
Institut za rudarstvo i metalurgiju, Bor.(2), 13-32.
https://doi.org/10.5937/mmeb1402013k

Kostić S, Vasović D, Okrajnov-Bajić R. New fresh concrete chemical admixture for tunnel lining design in the extreme winter conditions. in Mining and Metallurgy Engineering Bor. 2014;(2):13-32.
doi:10.5937/mmeb1402013k .

Kostić, Srđan, Vasović, Dejan, Okrajnov-Bajić, Ruža, "New fresh concrete chemical admixture for tunnel lining design in the extreme winter conditions" in Mining and Metallurgy Engineering Bor, no. 2 (2014):13-32,
https://doi.org/10.5937/mmeb1402013k . .

TY  - JOUR
AU  - Vasović, Dejan
AU  - Kostić, Srđan
AU  - Ravilić, Marina
AU  - Trajković, Slobodan
PY  - 2014
UR  - https://raf.arh.bg.ac.rs/handle/123456789/155
AB  - In present paper, the blast-induced ground motion and its effect on the neighboring structures are analyzed at the limestone quarry "Drenovac'' in central part of Serbia. Ground motion is examined by means of existing conventional predictors, with scaled distance as a main influential parameter, which gave satisfying prediction accuracy (R > 0.8), except in the case of Ambraseys-Hendron predictor. In the next step of the analysis, a feedforward three-layer back-propagation neural network is developed, with three input units (total charge, maximum charge per delay and distance from explosive charge to monitoring point) and only one output unit (peak particle velocity). The network is tested for the cases with different number of hidden nodes. The obtained results indicate that the model with six hidden nodes gives reasonable predictive precision (R approximate to 0.9), but with much lower values of mean-squared error in comparison to conventional predictors. In order to predict the influence level to the neighboring buildings, recorded peak particle velocities and frequency values were evaluated according to United States Bureau of Mines, USSR standard, German DIN4150, Australian standard, Indian DMGS circular 7 and Chinese safety regulations for blasting. Using the best conventional predictor, the relationship between the allowable amount of explosive and distance from explosive charge is determined for every vibration standard. Furthermore, the effect of air-blast overpressure is analyzed according to domestic regulations, with construction of a blasting chart for the permissible amount of explosive as a function of distance, for the allowable value of air-blast overpressure (200 Pa). The performed analysis indicates only small number of recordings above the upper allowable limit according to DIN4150 and DMGS standard, while, for all other vibration codes the registered values of ground velocity are within the permissible limits. As for the air-blast overpressure, no damage is expected to occur.
PB  - Springer Verlag
T2  - Environmental earth sciences
T1  - Environmental impact of blasting at Drenovac limestone quarry (Serbia)
VL  - 72
IS  - 10
SP  - 3915
EP  - 3928
DO  - 10.1007/s12665-014-3280-z
ER  - 

@article{
author = "Vasović, Dejan and Kostić, Srđan and Ravilić, Marina and Trajković, Slobodan",
year = "2014",
abstract = "In present paper, the blast-induced ground motion and its effect on the neighboring structures are analyzed at the limestone quarry "Drenovac'' in central part of Serbia. Ground motion is examined by means of existing conventional predictors, with scaled distance as a main influential parameter, which gave satisfying prediction accuracy (R > 0.8), except in the case of Ambraseys-Hendron predictor. In the next step of the analysis, a feedforward three-layer back-propagation neural network is developed, with three input units (total charge, maximum charge per delay and distance from explosive charge to monitoring point) and only one output unit (peak particle velocity). The network is tested for the cases with different number of hidden nodes. The obtained results indicate that the model with six hidden nodes gives reasonable predictive precision (R approximate to 0.9), but with much lower values of mean-squared error in comparison to conventional predictors. In order to predict the influence level to the neighboring buildings, recorded peak particle velocities and frequency values were evaluated according to United States Bureau of Mines, USSR standard, German DIN4150, Australian standard, Indian DMGS circular 7 and Chinese safety regulations for blasting. Using the best conventional predictor, the relationship between the allowable amount of explosive and distance from explosive charge is determined for every vibration standard. Furthermore, the effect of air-blast overpressure is analyzed according to domestic regulations, with construction of a blasting chart for the permissible amount of explosive as a function of distance, for the allowable value of air-blast overpressure (200 Pa). The performed analysis indicates only small number of recordings above the upper allowable limit according to DIN4150 and DMGS standard, while, for all other vibration codes the registered values of ground velocity are within the permissible limits. As for the air-blast overpressure, no damage is expected to occur.",
publisher = "Springer Verlag",
journal = "Environmental earth sciences",
title = "Environmental impact of blasting at Drenovac limestone quarry (Serbia)",
volume = "72",
number = "10",
pages = "3915-3928",
doi = "10.1007/s12665-014-3280-z"
}

Vasović, D., Kostić, S., Ravilić, M.,& Trajković, S.. (2014). Environmental impact of blasting at Drenovac limestone quarry (Serbia). in Environmental earth sciences
Springer Verlag., 72(10), 3915-3928.
https://doi.org/10.1007/s12665-014-3280-z

Vasović D, Kostić S, Ravilić M, Trajković S. Environmental impact of blasting at Drenovac limestone quarry (Serbia). in Environmental earth sciences. 2014;72(10):3915-3928.
doi:10.1007/s12665-014-3280-z .

Vasović, Dejan, Kostić, Srđan, Ravilić, Marina, Trajković, Slobodan, "Environmental impact of blasting at Drenovac limestone quarry (Serbia)" in Environmental earth sciences, 72, no. 10 (2014):3915-3928,
https://doi.org/10.1007/s12665-014-3280-z . .