TY  - JOUR
AU  - Malinović-Milićević, Slavica
AU  - Radovanović, Milan M.
AU  - Radenković, Sonja D.
AU  - Vyklyuk, Yaroslav
AU  - Milovanović, Boško
AU  - Milanović Pešić, Ana
AU  - Milenković, Milan
AU  - Popović, Vladimir
AU  - Petrović, Marko
AU  - Sydor, Petro
AU  - Gajić, Mirjana
PY  - 2023
UR  - https://dais.sanu.ac.rs/123456789/14049
AB  - This research is devoted to the determination of hidden dependencies between the flow of particles that come from the Sun and precipitation-induced floods in the United Kingdom (UK). The analysis covers 20 flood events during the period from October 2001 to December 2019. The parameters of solar activity were used as model input data, while precipitations data in the period 10 days before and during each flood event were used as model output. The time lag of 0–9 days was taken into account in the research. Correlation analysis was conducted to determine the degree of randomness for the time series of input and output parameters. For establishing a potential causative link, machine learning classification predictive modeling was applied. Two approaches, the decision tree, and the random forest were used. We analyzed the accuracy of classification models forecast from 0 to 9 days in advance. It was found that the most important factors for flood forecasting are proton density with a time lag of 9, differential proton flux in the range of 310–580 keV, and ion temperature. Research in this paper has shown that the decision tree model is more accurate and adequate in predicting the appearance of precipitation-induced floods up to 9 days ahead with an accuracy of 91%. The results of this study confirmed that by increasing technical capabilities, using improved machine learning techniques and large data sets, it is possible to improve the understanding of the physical link between the solar wind and tropospheric weather and help improve severe weather forecasting.
PB  - MDPI AG
T2  - Mathematics
T1  - Application of Solar Activity Time Series in Machine Learning Predictive Modeling of Precipitation-Induced Floods
SP  - 795
VL  - 11
IS  - 4
DO  - 10.3390/math11040795
UR  - https://hdl.handle.net/21.15107/rcub_dais_14049
ER  - 

@article{
author = "Malinović-Milićević, Slavica and Radovanović, Milan M. and Radenković, Sonja D. and Vyklyuk, Yaroslav and Milovanović, Boško and Milanović Pešić, Ana and Milenković, Milan and Popović, Vladimir and Petrović, Marko and Sydor, Petro and Gajić, Mirjana",
year = "2023",
abstract = "This research is devoted to the determination of hidden dependencies between the flow of particles that come from the Sun and precipitation-induced floods in the United Kingdom (UK). The analysis covers 20 flood events during the period from October 2001 to December 2019. The parameters of solar activity were used as model input data, while precipitations data in the period 10 days before and during each flood event were used as model output. The time lag of 0–9 days was taken into account in the research. Correlation analysis was conducted to determine the degree of randomness for the time series of input and output parameters. For establishing a potential causative link, machine learning classification predictive modeling was applied. Two approaches, the decision tree, and the random forest were used. We analyzed the accuracy of classification models forecast from 0 to 9 days in advance. It was found that the most important factors for flood forecasting are proton density with a time lag of 9, differential proton flux in the range of 310–580 keV, and ion temperature. Research in this paper has shown that the decision tree model is more accurate and adequate in predicting the appearance of precipitation-induced floods up to 9 days ahead with an accuracy of 91%. The results of this study confirmed that by increasing technical capabilities, using improved machine learning techniques and large data sets, it is possible to improve the understanding of the physical link between the solar wind and tropospheric weather and help improve severe weather forecasting.",
publisher = "MDPI AG",
journal = "Mathematics",
title = "Application of Solar Activity Time Series in Machine Learning Predictive Modeling of Precipitation-Induced Floods",
pages = "795",
volume = "11",
number = "4",
doi = "10.3390/math11040795",
url = "https://hdl.handle.net/21.15107/rcub_dais_14049"
}

Malinović-Milićević, S., Radovanović, M. M., Radenković, S. D., Vyklyuk, Y., Milovanović, B., Milanović Pešić, A., Milenković, M., Popović, V., Petrović, M., Sydor, P.,& Gajić, M.. (2023). Application of Solar Activity Time Series in Machine Learning Predictive Modeling of Precipitation-Induced Floods. in Mathematics
MDPI AG., 11(4), 795.
https://doi.org/10.3390/math11040795
https://hdl.handle.net/21.15107/rcub_dais_14049

Malinović-Milićević S, Radovanović MM, Radenković SD, Vyklyuk Y, Milovanović B, Milanović Pešić A, Milenković M, Popović V, Petrović M, Sydor P, Gajić M. Application of Solar Activity Time Series in Machine Learning Predictive Modeling of Precipitation-Induced Floods. in Mathematics. 2023;11(4):795.
doi:10.3390/math11040795
https://hdl.handle.net/21.15107/rcub_dais_14049 .

Malinović-Milićević, Slavica, Radovanović, Milan M., Radenković, Sonja D., Vyklyuk, Yaroslav, Milovanović, Boško, Milanović Pešić, Ana, Milenković, Milan, Popović, Vladimir, Petrović, Marko, Sydor, Petro, Gajić, Mirjana, "Application of Solar Activity Time Series in Machine Learning Predictive Modeling of Precipitation-Induced Floods" in Mathematics, 11, no. 4 (2023):795,
https://doi.org/10.3390/math11040795 .,
https://hdl.handle.net/21.15107/rcub_dais_14049 .

TY  - JOUR
AU  - Malinović-Milićević, Slavica
AU  - Vyklyuk, Yaroslav
AU  - Stanojević, Gorica
AU  - Radovanović, Milan M.
AU  - Doljak, Dejan
AU  - Ćurčić, Nina B.
PY  - 2021
UR  - https://dais.sanu.ac.rs/123456789/13304
AB  - In this paper, we described generation and performances of feedforward neural network model that could be used for a day ahead predictions of the daily maximum 1-h ozone concentration (1hO3) and 8-h average ozone concentration (8hO3) at one traffic and one background station in the urban area of Novi Sad,
Serbia. The six meteorological variables for the day preceding the forecast and forecast day, ozone concentrations in the day preceding the forecast, the number of the day of the year, and the number of the weekday for which ozone prediction was performed were utilized as inputs. The three-layer perceptron neural network models with the best performance were chosen by testing with different numbers of neurons in the hidden layer and different activation functions. The mean bias error, mean absolute error, root mean squared error, correlation coefficient, and index of agreement or Willmott’s Index for the validation data for 1hO3 forecasting were 0.005 μg m−3, 12.149 μg m−3, 15.926 μg m−3, 0.988, and 0.950, respectively, for the traffic station (Dnevnik), and − 0.565 μg m−3, 10.101 μg m−3, 12.962 μg m−3, 0.911, and 0.953, respectively, for the background station (Liman). For 8hO3 forecasting, statistical indicators were − 1.126 μg m−3, 10.614 μg m−3, 12.962 μg m−3, 0.910, and 0.948 respectively for the
station Dnevnik and − 0.001 μg m−3, 8.574 μg m−3, 10.741 μg m−3, 0.936, and 0.966, respectively, for the station Liman. According to the Kolmogorov–Smirnov test, there is no significant difference between measured and predicted data. Models showed a good performance in forecasting days with the high values over a certain threshold.
PB  - Switzerland : Springer Nature.
T2  - Environmental Monitoring and Assessment
T1  - Prediction of tropospheric ozone concentration using artificial neural networks at traffic and background urban locations in Novi Sad, Serbia
VL  - 193
IS  - 84
DO  - 10.1007/s10661-020-08821-1
UR  - https://hdl.handle.net/21.15107/rcub_dais_13304
ER  - 

@article{
author = "Malinović-Milićević, Slavica and Vyklyuk, Yaroslav and Stanojević, Gorica and Radovanović, Milan M. and Doljak, Dejan and Ćurčić, Nina B.",
year = "2021",
abstract = "In this paper, we described generation and performances of feedforward neural network model that could be used for a day ahead predictions of the daily maximum 1-h ozone concentration (1hO3) and 8-h average ozone concentration (8hO3) at one traffic and one background station in the urban area of Novi Sad,
Serbia. The six meteorological variables for the day preceding the forecast and forecast day, ozone concentrations in the day preceding the forecast, the number of the day of the year, and the number of the weekday for which ozone prediction was performed were utilized as inputs. The three-layer perceptron neural network models with the best performance were chosen by testing with different numbers of neurons in the hidden layer and different activation functions. The mean bias error, mean absolute error, root mean squared error, correlation coefficient, and index of agreement or Willmott’s Index for the validation data for 1hO3 forecasting were 0.005 μg m−3, 12.149 μg m−3, 15.926 μg m−3, 0.988, and 0.950, respectively, for the traffic station (Dnevnik), and − 0.565 μg m−3, 10.101 μg m−3, 12.962 μg m−3, 0.911, and 0.953, respectively, for the background station (Liman). For 8hO3 forecasting, statistical indicators were − 1.126 μg m−3, 10.614 μg m−3, 12.962 μg m−3, 0.910, and 0.948 respectively for the
station Dnevnik and − 0.001 μg m−3, 8.574 μg m−3, 10.741 μg m−3, 0.936, and 0.966, respectively, for the station Liman. According to the Kolmogorov–Smirnov test, there is no significant difference between measured and predicted data. Models showed a good performance in forecasting days with the high values over a certain threshold.",
publisher = "Switzerland : Springer Nature.",
journal = "Environmental Monitoring and Assessment",
title = "Prediction of tropospheric ozone concentration using artificial neural networks at traffic and background urban locations in Novi Sad, Serbia",
volume = "193",
number = "84",
doi = "10.1007/s10661-020-08821-1",
url = "https://hdl.handle.net/21.15107/rcub_dais_13304"
}

Malinović-Milićević, S., Vyklyuk, Y., Stanojević, G., Radovanović, M. M., Doljak, D.,& Ćurčić, N. B.. (2021). Prediction of tropospheric ozone concentration using artificial neural networks at traffic and background urban locations in Novi Sad, Serbia. in Environmental Monitoring and Assessment
Switzerland : Springer Nature.., 193(84).
https://doi.org/10.1007/s10661-020-08821-1
https://hdl.handle.net/21.15107/rcub_dais_13304

Malinović-Milićević S, Vyklyuk Y, Stanojević G, Radovanović MM, Doljak D, Ćurčić NB. Prediction of tropospheric ozone concentration using artificial neural networks at traffic and background urban locations in Novi Sad, Serbia. in Environmental Monitoring and Assessment. 2021;193(84).
doi:10.1007/s10661-020-08821-1
https://hdl.handle.net/21.15107/rcub_dais_13304 .

Malinović-Milićević, Slavica, Vyklyuk, Yaroslav, Stanojević, Gorica, Radovanović, Milan M., Doljak, Dejan, Ćurčić, Nina B., "Prediction of tropospheric ozone concentration using artificial neural networks at traffic and background urban locations in Novi Sad, Serbia" in Environmental Monitoring and Assessment, 193, no. 84 (2021),
https://doi.org/10.1007/s10661-020-08821-1 .,
https://hdl.handle.net/21.15107/rcub_dais_13304 .

TY  - JOUR
AU  - Vyklyuk, Yaroslav
AU  - Radovanović, Milan M.
AU  - Petrović, Marko D.
AU  - Ćurčić, Nina B.
AU  - Milenković, Milan
AU  - Malinović-Milićević, Slavica
AU  - Milovanović, Bosko
AU  - Yamashkin, Anatoliy A.
AU  - Milanović Pešić, Ana
AU  - Lukić, Dobrila
AU  - Gajić, Mirjana
PY  - 2020
UR  - https://dais.sanu.ac.rs/123456789/13407
AB  - The impact of solar activity on environmental processes is difficult to understand and complex for empirical modeling. This study aimed to establish forecast models of the meteorological conditions in the forest fire areas based on the solar activity parameters applying the neural networks approach. During July and August 2018, severe forest fires simultaneously occurred in the State of California (USA), Portugal, and Greece. Air temperature and humidity data together with solar parameters (integral flux of solar protons, differential electron flux and proton flux, solar wind plasma parameters, and solar radio flux at 10.7 cm data) were used in long short-term memory (LSTM) recurrent neural network ensembles. It is found that solar activity mostly affects the humidity for two stations in California and Portugal (an increase in the integral flux of solar protons of > 30 MeV by 10% increases the humidity by 3.25%, 1.65%, and 1.57%, respectively). Furthermore, an increase in air temperature of 10% increases the humidity by 2.55%, 2.01%, and 0.26%, respectively. It is shown that temperature is less sensitive to changes in solar parameters but depends on previous conditions (previous increase of 10% increases the current temperature by 0.75%, 0.34%, and 0.33%, respectively). Humidity in Greece is mostly impacted by solar flux F10.7 cm and previous values of humidity. An increase in these factors by 10% will lead to a decrease in the humidity of 3.89% or an increase of 1.31%, while air temperature mostly depends on ion temperature. If this factor increases by 10%, it will lead to air temperature rising by 0.42%
PB  - Basel, Switzerland : MDPI.
T2  - Sustainability
T1  - Connection of Solar Activities and Forest Fires in 2018: Events in the USA (California), Portugal and Greece
SP  - 10261
VL  - 12
IS  - 24
DO  - 10.3390/su122410261
UR  - https://hdl.handle.net/21.15107/rcub_dais_13407
ER  - 

@article{
author = "Vyklyuk, Yaroslav and Radovanović, Milan M. and Petrović, Marko D. and Ćurčić, Nina B. and Milenković, Milan and Malinović-Milićević, Slavica and Milovanović, Bosko and Yamashkin, Anatoliy A. and Milanović Pešić, Ana and Lukić, Dobrila and Gajić, Mirjana",
year = "2020",
abstract = "The impact of solar activity on environmental processes is difficult to understand and complex for empirical modeling. This study aimed to establish forecast models of the meteorological conditions in the forest fire areas based on the solar activity parameters applying the neural networks approach. During July and August 2018, severe forest fires simultaneously occurred in the State of California (USA), Portugal, and Greece. Air temperature and humidity data together with solar parameters (integral flux of solar protons, differential electron flux and proton flux, solar wind plasma parameters, and solar radio flux at 10.7 cm data) were used in long short-term memory (LSTM) recurrent neural network ensembles. It is found that solar activity mostly affects the humidity for two stations in California and Portugal (an increase in the integral flux of solar protons of > 30 MeV by 10% increases the humidity by 3.25%, 1.65%, and 1.57%, respectively). Furthermore, an increase in air temperature of 10% increases the humidity by 2.55%, 2.01%, and 0.26%, respectively. It is shown that temperature is less sensitive to changes in solar parameters but depends on previous conditions (previous increase of 10% increases the current temperature by 0.75%, 0.34%, and 0.33%, respectively). Humidity in Greece is mostly impacted by solar flux F10.7 cm and previous values of humidity. An increase in these factors by 10% will lead to a decrease in the humidity of 3.89% or an increase of 1.31%, while air temperature mostly depends on ion temperature. If this factor increases by 10%, it will lead to air temperature rising by 0.42%",
publisher = "Basel, Switzerland : MDPI.",
journal = "Sustainability",
title = "Connection of Solar Activities and Forest Fires in 2018: Events in the USA (California), Portugal and Greece",
pages = "10261",
volume = "12",
number = "24",
doi = "10.3390/su122410261",
url = "https://hdl.handle.net/21.15107/rcub_dais_13407"
}

Vyklyuk, Y., Radovanović, M. M., Petrović, M. D., Ćurčić, N. B., Milenković, M., Malinović-Milićević, S., Milovanović, B., Yamashkin, A. A., Milanović Pešić, A., Lukić, D.,& Gajić, M.. (2020). Connection of Solar Activities and Forest Fires in 2018: Events in the USA (California), Portugal and Greece. in Sustainability
Basel, Switzerland : MDPI.., 12(24), 10261.
https://doi.org/10.3390/su122410261
https://hdl.handle.net/21.15107/rcub_dais_13407

Vyklyuk Y, Radovanović MM, Petrović MD, Ćurčić NB, Milenković M, Malinović-Milićević S, Milovanović B, Yamashkin AA, Milanović Pešić A, Lukić D, Gajić M. Connection of Solar Activities and Forest Fires in 2018: Events in the USA (California), Portugal and Greece. in Sustainability. 2020;12(24):10261.
doi:10.3390/su122410261
https://hdl.handle.net/21.15107/rcub_dais_13407 .

Vyklyuk, Yaroslav, Radovanović, Milan M., Petrović, Marko D., Ćurčić, Nina B., Milenković, Milan, Malinović-Milićević, Slavica, Milovanović, Bosko, Yamashkin, Anatoliy A., Milanović Pešić, Ana, Lukić, Dobrila, Gajić, Mirjana, "Connection of Solar Activities and Forest Fires in 2018: Events in the USA (California), Portugal and Greece" in Sustainability, 12, no. 24 (2020):10261,
https://doi.org/10.3390/su122410261 .,
https://hdl.handle.net/21.15107/rcub_dais_13407 .

TY  - JOUR
AU  - Radovanović, Milan M.
AU  - Vyklyuk, Yaroslav
AU  - Stevančević, Milan T.
AU  - Milenković, Milan Đ
AU  - Jakovljević, Dejana M.
AU  - Petrović, Marko D.
AU  - Malinović-Milićević, Slavica B.
AU  - Vuković, Natalia
AU  - Vujko, Aleksandra Đ.
AU  - Sydor
AU  - Yamashkin, Anatoliy
AU  - Sydor, Petro
AU  - Vuković, Darko B.
AU  - Škoda, Miroslav
PY  - 2019
UR  - https://dais.sanu.ac.rs/123456789/13409
AB  - Forest fires that occurred in Portugal on June 18, 2017, caused several tens of
human casualties. The cause of their emergence, as well as many others that occurred in western Europe at the same time remained unknown. Taking into account consequences, including loss of human lives and endangerment of ecosystem sustainability, discovering of the forest fires causes is the very significant question. The heliocentric hypothesis has indirectly been tested, according to which charged particles are a possible cause of forest fires. We must point out that it was not possible to verify whether in this specific case the particles by reaching the ground and burning the plant mass create the initial phase of the formation of the flame. Therefore, we have tried to determine whether during the critical period, i. e. from June 15-19 there is a certain statistical connection between certain parameters of the solar wind and meteorological elements. Based on the hourly values of the charged particles flow, a correlation analysis was performed with hourly values of individual meteorological elements including time lag at Monte Real station. The application of the Adaptive Neuro Fuzzy Inference System models has shown that there is a high degree of connection between the flow of protons and the analyzed meteorological elements in Portugal. However, further verification of this hypothesis requires further laboratory testing.
PB  - Belgrade :  Vinča Institute of Nuclear Sciences
T2  - Thermal Science
T1  - Forest fires in Portugal - case study, 18 june 2017
SP  - 73
EP  - 86
VL  - 23
IS  - 1
DO  - 10.2298/TSCI180803251R
UR  - https://hdl.handle.net/21.15107/rcub_dais_13409
ER  - 

@article{
author = "Radovanović, Milan M. and Vyklyuk, Yaroslav and Stevančević, Milan T. and Milenković, Milan Đ and Jakovljević, Dejana M. and Petrović, Marko D. and Malinović-Milićević, Slavica B. and Vuković, Natalia and Vujko, Aleksandra Đ. and Sydor and Yamashkin, Anatoliy and Sydor, Petro and Vuković, Darko B. and Škoda, Miroslav",
year = "2019",
abstract = "Forest fires that occurred in Portugal on June 18, 2017, caused several tens of
human casualties. The cause of their emergence, as well as many others that occurred in western Europe at the same time remained unknown. Taking into account consequences, including loss of human lives and endangerment of ecosystem sustainability, discovering of the forest fires causes is the very significant question. The heliocentric hypothesis has indirectly been tested, according to which charged particles are a possible cause of forest fires. We must point out that it was not possible to verify whether in this specific case the particles by reaching the ground and burning the plant mass create the initial phase of the formation of the flame. Therefore, we have tried to determine whether during the critical period, i. e. from June 15-19 there is a certain statistical connection between certain parameters of the solar wind and meteorological elements. Based on the hourly values of the charged particles flow, a correlation analysis was performed with hourly values of individual meteorological elements including time lag at Monte Real station. The application of the Adaptive Neuro Fuzzy Inference System models has shown that there is a high degree of connection between the flow of protons and the analyzed meteorological elements in Portugal. However, further verification of this hypothesis requires further laboratory testing.",
publisher = "Belgrade :  Vinča Institute of Nuclear Sciences",
journal = "Thermal Science",
title = "Forest fires in Portugal - case study, 18 june 2017",
pages = "73-86",
volume = "23",
number = "1",
doi = "10.2298/TSCI180803251R",
url = "https://hdl.handle.net/21.15107/rcub_dais_13409"
}

Radovanović, M. M., Vyklyuk, Y., Stevančević, M. T., Milenković, M. Đ., Jakovljević, D. M., Petrović, M. D., Malinović-Milićević, S. B., Vuković, N., Vujko, A. Đ., Sydor, Yamashkin, A., Sydor, P., Vuković, D. B.,& Škoda, M.. (2019). Forest fires in Portugal - case study, 18 june 2017. in Thermal Science
Belgrade :  Vinča Institute of Nuclear Sciences., 23(1), 73-86.
https://doi.org/10.2298/TSCI180803251R
https://hdl.handle.net/21.15107/rcub_dais_13409

Radovanović MM, Vyklyuk Y, Stevančević MT, Milenković MĐ, Jakovljević DM, Petrović MD, Malinović-Milićević SB, Vuković N, Vujko AĐ, Sydor, Yamashkin A, Sydor P, Vuković DB, Škoda M. Forest fires in Portugal - case study, 18 june 2017. in Thermal Science. 2019;23(1):73-86.
doi:10.2298/TSCI180803251R
https://hdl.handle.net/21.15107/rcub_dais_13409 .

Radovanović, Milan M., Vyklyuk, Yaroslav, Stevančević, Milan T., Milenković, Milan Đ, Jakovljević, Dejana M., Petrović, Marko D., Malinović-Milićević, Slavica B., Vuković, Natalia, Vujko, Aleksandra Đ., Sydor, Yamashkin, Anatoliy, Sydor, Petro, Vuković, Darko B., Škoda, Miroslav, "Forest fires in Portugal - case study, 18 june 2017" in Thermal Science, 23, no. 1 (2019):73-86,
https://doi.org/10.2298/TSCI180803251R .,
https://hdl.handle.net/21.15107/rcub_dais_13409 .

TY  - JOUR
AU  - Vyklyuk, Yaroslav
AU  - Radovanović, Milan M.
AU  - Milovanović, Boško
AU  - Milenković, Milan
AU  - Petrović, Marko
AU  - Doljak, Dejan
AU  - Malinović-Milićević, Slavica
AU  - Vuković, Natalia
AU  - Vujko, Aleksandra
AU  - Matsiuk, Nataliia
AU  - Mukherjee, Saumitra
PY  - 2019
UR  - https://dais.sanu.ac.rs/123456789/13408
AB  - This research is devoted to the determination of the causal relationship between the flow of particles that are coming from the Sun and the hurricanes Irma, Jose, and Katia. To accomplish this, the lag correlation analysis was performed. High correlation coefficients confirmed a preliminary conclusion about the relationship between solar activities and the hurricane phenomenon, which allows further research. Five parameters i.e. characteristics of solar activity (10.7 cm solar radio flux (F10.7), the flows of protons and electrons with maximum energy, speed and density of solar wind particles) were chosen as model input, while the wind speed and air pressure of Irma, Jose, and Katia hurricanes were used as model output. Input data were sampled to a six hours interval in order to adapt the time interval to the observed data about hurricanes, in the period between September 28 and December 21, 2017. As a result of the preliminary analysis, using 12,274,264 linear models by parallel calculations, six of them were chosen as best. The identified lags were the basis for refinement of models with the artificial neural networks. Multilayer perceptrons with back propagation and recurrent LSTM have been chosen as commonly used artificial neural networks. Comparison of the accuracy of both linear and artificial neural networks results confirmed the adequacy of these models and made it possible to take into account the dynamics of the solar wind. Sensitivity analysis has shown that F10.7 has the greatest impact on the wind speed of the hurricanes. Despite low sensitivity of pressure to change the parameters of the solar wind, their strong fluctuations can cause a sharp decrease in pressure, and therefore the appearance of hurricanes.
PB  - Switzerland : Springer Nature
T2  - Astrophysics and Space Science
T1  - Space weather and hurricanes Irma, Jose and Katia
VL  - 364
IS  - 154
DO  - 10.1007/s10509-019-3646-5
UR  - https://hdl.handle.net/21.15107/rcub_dais_13408
ER  - 

@article{
author = "Vyklyuk, Yaroslav and Radovanović, Milan M. and Milovanović, Boško and Milenković, Milan and Petrović, Marko and Doljak, Dejan and Malinović-Milićević, Slavica and Vuković, Natalia and Vujko, Aleksandra and Matsiuk, Nataliia and Mukherjee, Saumitra",
year = "2019",
abstract = "This research is devoted to the determination of the causal relationship between the flow of particles that are coming from the Sun and the hurricanes Irma, Jose, and Katia. To accomplish this, the lag correlation analysis was performed. High correlation coefficients confirmed a preliminary conclusion about the relationship between solar activities and the hurricane phenomenon, which allows further research. Five parameters i.e. characteristics of solar activity (10.7 cm solar radio flux (F10.7), the flows of protons and electrons with maximum energy, speed and density of solar wind particles) were chosen as model input, while the wind speed and air pressure of Irma, Jose, and Katia hurricanes were used as model output. Input data were sampled to a six hours interval in order to adapt the time interval to the observed data about hurricanes, in the period between September 28 and December 21, 2017. As a result of the preliminary analysis, using 12,274,264 linear models by parallel calculations, six of them were chosen as best. The identified lags were the basis for refinement of models with the artificial neural networks. Multilayer perceptrons with back propagation and recurrent LSTM have been chosen as commonly used artificial neural networks. Comparison of the accuracy of both linear and artificial neural networks results confirmed the adequacy of these models and made it possible to take into account the dynamics of the solar wind. Sensitivity analysis has shown that F10.7 has the greatest impact on the wind speed of the hurricanes. Despite low sensitivity of pressure to change the parameters of the solar wind, their strong fluctuations can cause a sharp decrease in pressure, and therefore the appearance of hurricanes.",
publisher = "Switzerland : Springer Nature",
journal = "Astrophysics and Space Science",
title = "Space weather and hurricanes Irma, Jose and Katia",
volume = "364",
number = "154",
doi = "10.1007/s10509-019-3646-5",
url = "https://hdl.handle.net/21.15107/rcub_dais_13408"
}

Vyklyuk, Y., Radovanović, M. M., Milovanović, B., Milenković, M., Petrović, M., Doljak, D., Malinović-Milićević, S., Vuković, N., Vujko, A., Matsiuk, N.,& Mukherjee, S.. (2019). Space weather and hurricanes Irma, Jose and Katia. in Astrophysics and Space Science
Switzerland : Springer Nature., 364(154).
https://doi.org/10.1007/s10509-019-3646-5
https://hdl.handle.net/21.15107/rcub_dais_13408

Vyklyuk Y, Radovanović MM, Milovanović B, Milenković M, Petrović M, Doljak D, Malinović-Milićević S, Vuković N, Vujko A, Matsiuk N, Mukherjee S. Space weather and hurricanes Irma, Jose and Katia. in Astrophysics and Space Science. 2019;364(154).
doi:10.1007/s10509-019-3646-5
https://hdl.handle.net/21.15107/rcub_dais_13408 .

Vyklyuk, Yaroslav, Radovanović, Milan M., Milovanović, Boško, Milenković, Milan, Petrović, Marko, Doljak, Dejan, Malinović-Milićević, Slavica, Vuković, Natalia, Vujko, Aleksandra, Matsiuk, Nataliia, Mukherjee, Saumitra, "Space weather and hurricanes Irma, Jose and Katia" in Astrophysics and Space Science, 364, no. 154 (2019),
https://doi.org/10.1007/s10509-019-3646-5 .,
https://hdl.handle.net/21.15107/rcub_dais_13408 .

TY  - CONF
AU  - Vyklyuk, Yaroslav
AU  - Radovanović, Milan M.
AU  - Malinović-Milićević, Slavica
PY  - 2019
UR  - https://dais.sanu.ac.rs/123456789/13404
PB  - Belgrade : Geographical Institute "Jovan Cvijić" SASA
C3  - Integrations of satellite and ground-based observations and multi-disciplinarity in research and prediction of different types of hazards in solar system: book of abstracts
T1  - DEEP Learning LSTM recurrent neural network for consequence forecasting of the solar wind disturbance
SP  - 24
EP  - 25
UR  - https://hdl.handle.net/21.15107/rcub_dais_13404
ER  - 

@conference{
author = "Vyklyuk, Yaroslav and Radovanović, Milan M. and Malinović-Milićević, Slavica",
year = "2019",
publisher = "Belgrade : Geographical Institute "Jovan Cvijić" SASA",
journal = "Integrations of satellite and ground-based observations and multi-disciplinarity in research and prediction of different types of hazards in solar system: book of abstracts",
title = "DEEP Learning LSTM recurrent neural network for consequence forecasting of the solar wind disturbance",
pages = "24-25",
url = "https://hdl.handle.net/21.15107/rcub_dais_13404"
}

Vyklyuk, Y., Radovanović, M. M.,& Malinović-Milićević, S.. (2019). DEEP Learning LSTM recurrent neural network for consequence forecasting of the solar wind disturbance. in Integrations of satellite and ground-based observations and multi-disciplinarity in research and prediction of different types of hazards in solar system: book of abstracts
Belgrade : Geographical Institute "Jovan Cvijić" SASA., 24-25.
https://hdl.handle.net/21.15107/rcub_dais_13404

Vyklyuk Y, Radovanović MM, Malinović-Milićević S. DEEP Learning LSTM recurrent neural network for consequence forecasting of the solar wind disturbance. in Integrations of satellite and ground-based observations and multi-disciplinarity in research and prediction of different types of hazards in solar system: book of abstracts. 2019;:24-25.
https://hdl.handle.net/21.15107/rcub_dais_13404 .

Vyklyuk, Yaroslav, Radovanović, Milan M., Malinović-Milićević, Slavica, "DEEP Learning LSTM recurrent neural network for consequence forecasting of the solar wind disturbance" in Integrations of satellite and ground-based observations and multi-disciplinarity in research and prediction of different types of hazards in solar system: book of abstracts (2019):24-25,
https://hdl.handle.net/21.15107/rcub_dais_13404 .

TY  - JOUR
AU  - Malinovic-Milicevic, Slavica
AU  - Vyklyuk, Yaroslav
AU  - Radovanovic, Milan M.
AU  - Petrovic, Marko D
PY  - 2018
UR  - https://dais.sanu.ac.rs/123456789/12919
AB  - In this article, we proposed a simple neural network (NN) technique for estimating
erythemal ultraviolet (UV) radiation in Novi Sad (Serbia) using available input
parameters. The technique implies the use of one of two models depending on the
availability of the input parameter: (a) NN model 1 (NNM1) which uses global
solar radiation, clearness index, cloudiness and air mass; and (b) NN model
2 (NNM2) which adds total ozone content (TOC) to the NNM1 inputs. The three
feed-forward NNs with different internal structures and back propagation learning
method for each NN model were used in modelling. The parallel calculation was
used for learning each NN. The results showed that the NNM1 provides satisfactory estimate (R = 0.975, MBE = −0.614%, MAPE = 12.580%, RMSE =
17.716%) and that additional use of TOC NNM2 considerably improves the
results (R = 0.982, MBE = −0.726%, MAPE = 10.161%, RMSE = 14.509%).
The performances of developed NNMs become significantly better if warm part of
the year is isolated (MAPE = 10.981 and 8.958; RMSE = 13.889 and 11.709, for
NNM1 and NNM2, respectively). Variations of reconstructed annual averages of
daily doses in the period 1949–2012 indicate ability of the technique to model the
relationship between erythemal UV radiation and the affecting atmospheric factors. The analysis showed that the increasing trend during the warm part of the
year in the period 1981–1996 was mainly caused by TOC, while the increase after
1996 was to a greater extent caused by cloudiness.
PB  - United Kingdom : John Wiley & Sons Ltd.
T2  - International Journal of Climatology
T1  - Long-term erythemal ultraviolet radiation in Novi Sad (Serbia) reconstructed by neural network modelling
SP  - 3264
EP  - 3272
VL  - 38
IS  - 8
DO  - 10.1002/joc.5499
UR  - https://hdl.handle.net/21.15107/rcub_dais_12919
ER  - 

@article{
author = "Malinovic-Milicevic, Slavica and Vyklyuk, Yaroslav and Radovanovic, Milan M. and Petrovic, Marko D",
year = "2018",
abstract = "In this article, we proposed a simple neural network (NN) technique for estimating
erythemal ultraviolet (UV) radiation in Novi Sad (Serbia) using available input
parameters. The technique implies the use of one of two models depending on the
availability of the input parameter: (a) NN model 1 (NNM1) which uses global
solar radiation, clearness index, cloudiness and air mass; and (b) NN model
2 (NNM2) which adds total ozone content (TOC) to the NNM1 inputs. The three
feed-forward NNs with different internal structures and back propagation learning
method for each NN model were used in modelling. The parallel calculation was
used for learning each NN. The results showed that the NNM1 provides satisfactory estimate (R = 0.975, MBE = −0.614%, MAPE = 12.580%, RMSE =
17.716%) and that additional use of TOC NNM2 considerably improves the
results (R = 0.982, MBE = −0.726%, MAPE = 10.161%, RMSE = 14.509%).
The performances of developed NNMs become significantly better if warm part of
the year is isolated (MAPE = 10.981 and 8.958; RMSE = 13.889 and 11.709, for
NNM1 and NNM2, respectively). Variations of reconstructed annual averages of
daily doses in the period 1949–2012 indicate ability of the technique to model the
relationship between erythemal UV radiation and the affecting atmospheric factors. The analysis showed that the increasing trend during the warm part of the
year in the period 1981–1996 was mainly caused by TOC, while the increase after
1996 was to a greater extent caused by cloudiness.",
publisher = "United Kingdom : John Wiley & Sons Ltd.",
journal = "International Journal of Climatology",
title = "Long-term erythemal ultraviolet radiation in Novi Sad (Serbia) reconstructed by neural network modelling",
pages = "3264-3272",
volume = "38",
number = "8",
doi = "10.1002/joc.5499",
url = "https://hdl.handle.net/21.15107/rcub_dais_12919"
}

Malinovic-Milicevic, S., Vyklyuk, Y., Radovanovic, M. M.,& Petrovic, M. D.. (2018). Long-term erythemal ultraviolet radiation in Novi Sad (Serbia) reconstructed by neural network modelling. in International Journal of Climatology
United Kingdom : John Wiley & Sons Ltd.., 38(8), 3264-3272.
https://doi.org/10.1002/joc.5499
https://hdl.handle.net/21.15107/rcub_dais_12919

Malinovic-Milicevic S, Vyklyuk Y, Radovanovic MM, Petrovic MD. Long-term erythemal ultraviolet radiation in Novi Sad (Serbia) reconstructed by neural network modelling. in International Journal of Climatology. 2018;38(8):3264-3272.
doi:10.1002/joc.5499
https://hdl.handle.net/21.15107/rcub_dais_12919 .

Malinovic-Milicevic, Slavica, Vyklyuk, Yaroslav, Radovanovic, Milan M., Petrovic, Marko D, "Long-term erythemal ultraviolet radiation in Novi Sad (Serbia) reconstructed by neural network modelling" in International Journal of Climatology, 38, no. 8 (2018):3264-3272,
https://doi.org/10.1002/joc.5499 .,
https://hdl.handle.net/21.15107/rcub_dais_12919 .

TY  - JOUR
AU  - Vyklyuk, Yaroslav
AU  - Radovanović, Milan. M.
AU  - Stanojević, Gorica B.
AU  - Milovanović, Boško
AU  - Taras, Leko
AU  - Milenković, Milan
AU  - Petrović, Marko
AU  - Yamashkin, Anatoly A.
AU  - Milanović Pešić, Ana
AU  - Jakovljević, Dejana
AU  - Malinović Milićević, Slavica
PY  - 2018
UR  - https://dais.sanu.ac.rs/123456789/13855
AB  - This research presents improved results on modelling relationship between the flow of charged particles that are coming from the Sun and hurricanes. For establishing eventual link, the methods of Big Data, such as Adaptive Neuro Fuzzy Inference System (ANFIS), Parallel Calculations, Fractal analysis etc., are applied. The parameters of solar activity were used as model input data, while data on hurricane phenomenon were used as model output, and both of these on daily level for May–October in period 1999–2013. The nonlinear R/S analysis was conducted to determine the degree of randomness for time series of input and output parameters. The time lag of 0–10 days was taken into account in the research. It led to growing input parameters up to 99. The problem of finding hidden dependencies in large databases refers to the problems of Data Mining. The ANFIS with Sugeno function of zero order was selected as a method of output fuzzy system. The “brute-force attack” method was used to find the most significant factors from all data. To do this, more than 3 million ANFIS models were tested on Computer Cluster using Parallel Calculation. Within the experiments, eight input factors were calculated as a base for building the final ANFIS models. These models can predict up to 39% of the hurricanes. This means, if causal link exists, approximately every third penetration of charged particles from coronary hole(s) or/and from the energetic region(s) toward the Earth precede the hurricanes.
PB  - Netherlands : Elsevier Ltd.
T2  - Journal of Atmospheric and Solar-Terrestrial Physics
T1  - Hurricane genesis modeling based on the relationship between solar activity and hurricanes II
SP  - 159
EP  - 164
VL  - 180
DO  - 10.1016/j.jastp.2017.09.008
UR  - https://hdl.handle.net/21.15107/rcub_dais_13855
ER  - 

@article{
author = "Vyklyuk, Yaroslav and Radovanović, Milan. M. and Stanojević, Gorica B. and Milovanović, Boško and Taras, Leko and Milenković, Milan and Petrović, Marko and Yamashkin, Anatoly A. and Milanović Pešić, Ana and Jakovljević, Dejana and Malinović Milićević, Slavica",
year = "2018",
abstract = "This research presents improved results on modelling relationship between the flow of charged particles that are coming from the Sun and hurricanes. For establishing eventual link, the methods of Big Data, such as Adaptive Neuro Fuzzy Inference System (ANFIS), Parallel Calculations, Fractal analysis etc., are applied. The parameters of solar activity were used as model input data, while data on hurricane phenomenon were used as model output, and both of these on daily level for May–October in period 1999–2013. The nonlinear R/S analysis was conducted to determine the degree of randomness for time series of input and output parameters. The time lag of 0–10 days was taken into account in the research. It led to growing input parameters up to 99. The problem of finding hidden dependencies in large databases refers to the problems of Data Mining. The ANFIS with Sugeno function of zero order was selected as a method of output fuzzy system. The “brute-force attack” method was used to find the most significant factors from all data. To do this, more than 3 million ANFIS models were tested on Computer Cluster using Parallel Calculation. Within the experiments, eight input factors were calculated as a base for building the final ANFIS models. These models can predict up to 39% of the hurricanes. This means, if causal link exists, approximately every third penetration of charged particles from coronary hole(s) or/and from the energetic region(s) toward the Earth precede the hurricanes.",
publisher = "Netherlands : Elsevier Ltd.",
journal = "Journal of Atmospheric and Solar-Terrestrial Physics",
title = "Hurricane genesis modeling based on the relationship between solar activity and hurricanes II",
pages = "159-164",
volume = "180",
doi = "10.1016/j.jastp.2017.09.008",
url = "https://hdl.handle.net/21.15107/rcub_dais_13855"
}

Vyklyuk, Y., Radovanović, Milan. M., Stanojević, G. B., Milovanović, B., Taras, L., Milenković, M., Petrović, M., Yamashkin, A. A., Milanović Pešić, A., Jakovljević, D.,& Malinović Milićević, S.. (2018). Hurricane genesis modeling based on the relationship between solar activity and hurricanes II. in Journal of Atmospheric and Solar-Terrestrial Physics
Netherlands : Elsevier Ltd.., 180, 159-164.
https://doi.org/10.1016/j.jastp.2017.09.008
https://hdl.handle.net/21.15107/rcub_dais_13855

Vyklyuk Y, Radovanović MM, Stanojević GB, Milovanović B, Taras L, Milenković M, Petrović M, Yamashkin AA, Milanović Pešić A, Jakovljević D, Malinović Milićević S. Hurricane genesis modeling based on the relationship between solar activity and hurricanes II. in Journal of Atmospheric and Solar-Terrestrial Physics. 2018;180:159-164.
doi:10.1016/j.jastp.2017.09.008
https://hdl.handle.net/21.15107/rcub_dais_13855 .

Vyklyuk, Yaroslav, Radovanović, Milan. M., Stanojević, Gorica B., Milovanović, Boško, Taras, Leko, Milenković, Milan, Petrović, Marko, Yamashkin, Anatoly A., Milanović Pešić, Ana, Jakovljević, Dejana, Malinović Milićević, Slavica, "Hurricane genesis modeling based on the relationship between solar activity and hurricanes II" in Journal of Atmospheric and Solar-Terrestrial Physics, 180 (2018):159-164,
https://doi.org/10.1016/j.jastp.2017.09.008 .,
https://hdl.handle.net/21.15107/rcub_dais_13855 .

TY  - JOUR
AU  - Vyklyuk, Yaroslav
AU  - Radovanović, Milan M.
AU  - Milovanović, Boško
AU  - Leko, Taras
AU  - Milenković, Milan
AU  - Milošević, Zoran
AU  - Milanović Pešić, Ana
AU  - Jakovljević, Dejana M.
PY  - 2017
UR  - https://dais.sanu.ac.rs/123456789/1058
AB  - The work examines the potential causative link between the flow of charged particles that are coming from the Sun and hurricanes. For establishing eventual link, the method of correlation analysis is applied, but with the phase shift of 0–5 days. There are nine parameters which are observed as an input, and daily values of the hurricane phenomenon are observed as an output in the period May–October 1999–2013. The results that have been obtained are considerably weak, leading to the need of applying the method of nonlinear analysis. The R/S analysis was conducted to determine the degree of randomness for time series of input and output parameters. The calculated Hurst index of variables X4–X9 is persistent (0.71–0.96). That allows us to conclude that the dynamics of these time series is heavily dependent on the same factors or on each other. Unlike the previous index, we have concluded that the behavior of high-energy protons (X1–X3) and the number of hurricane time series are completely stochastic. The problem of finding hidden dependencies in large databases refers to problems of data mining. The Sugeno function of zero order was selected as a method of output fuzzy system. Bearing in mind the available equipment, the models had to be shortened to the phase shift of 0–3 days. The “brute-force attack” method was used to find the most significant factors from all data. Within the experiments, six input factors were calculated which became the basis for building the final ANFIS models. These models can predict 22–26 % of the hurricanes.
PB  - Springer Netherlands
T2  - Natural Hazards
T1  - Hurricane genesis modelling based on the relationship between solar activity and hurricanes
SP  - 1043
EP  - 1062
VL  - 85
IS  - 2
DO  - 10.1007/s11069-016-2620-6
UR  - https://hdl.handle.net/21.15107/rcub_dais_1058
ER  - 

@article{
author = "Vyklyuk, Yaroslav and Radovanović, Milan M. and Milovanović, Boško and Leko, Taras and Milenković, Milan and Milošević, Zoran and Milanović Pešić, Ana and Jakovljević, Dejana M.",
year = "2017",
abstract = "The work examines the potential causative link between the flow of charged particles that are coming from the Sun and hurricanes. For establishing eventual link, the method of correlation analysis is applied, but with the phase shift of 0–5 days. There are nine parameters which are observed as an input, and daily values of the hurricane phenomenon are observed as an output in the period May–October 1999–2013. The results that have been obtained are considerably weak, leading to the need of applying the method of nonlinear analysis. The R/S analysis was conducted to determine the degree of randomness for time series of input and output parameters. The calculated Hurst index of variables X4–X9 is persistent (0.71–0.96). That allows us to conclude that the dynamics of these time series is heavily dependent on the same factors or on each other. Unlike the previous index, we have concluded that the behavior of high-energy protons (X1–X3) and the number of hurricane time series are completely stochastic. The problem of finding hidden dependencies in large databases refers to problems of data mining. The Sugeno function of zero order was selected as a method of output fuzzy system. Bearing in mind the available equipment, the models had to be shortened to the phase shift of 0–3 days. The “brute-force attack” method was used to find the most significant factors from all data. Within the experiments, six input factors were calculated which became the basis for building the final ANFIS models. These models can predict 22–26 % of the hurricanes.",
publisher = "Springer Netherlands",
journal = "Natural Hazards",
title = "Hurricane genesis modelling based on the relationship between solar activity and hurricanes",
pages = "1043-1062",
volume = "85",
number = "2",
doi = "10.1007/s11069-016-2620-6",
url = "https://hdl.handle.net/21.15107/rcub_dais_1058"
}

Vyklyuk, Y., Radovanović, M. M., Milovanović, B., Leko, T., Milenković, M., Milošević, Z., Milanović Pešić, A.,& Jakovljević, D. M.. (2017). Hurricane genesis modelling based on the relationship between solar activity and hurricanes. in Natural Hazards
Springer Netherlands., 85(2), 1043-1062.
https://doi.org/10.1007/s11069-016-2620-6
https://hdl.handle.net/21.15107/rcub_dais_1058

Vyklyuk Y, Radovanović MM, Milovanović B, Leko T, Milenković M, Milošević Z, Milanović Pešić A, Jakovljević DM. Hurricane genesis modelling based on the relationship between solar activity and hurricanes. in Natural Hazards. 2017;85(2):1043-1062.
doi:10.1007/s11069-016-2620-6
https://hdl.handle.net/21.15107/rcub_dais_1058 .

Vyklyuk, Yaroslav, Radovanović, Milan M., Milovanović, Boško, Leko, Taras, Milenković, Milan, Milošević, Zoran, Milanović Pešić, Ana, Jakovljević, Dejana M., "Hurricane genesis modelling based on the relationship between solar activity and hurricanes" in Natural Hazards, 85, no. 2 (2017):1043-1062,
https://doi.org/10.1007/s11069-016-2620-6 .,
https://hdl.handle.net/21.15107/rcub_dais_1058 .

TY  - JOUR
AU  - Radovanović, Milan M.
AU  - Vyklyuk, Yaroslav
AU  - Malinović-Milićević, Slavica B.
AU  - Jakovljević, Dejana M.
AU  - Pecelj, Milica R.
PY  - 2015
UR  - https://dais.sanu.ac.rs/123456789/13498
AB  - The work examines the potential causative link between the flow of protons, i. e. temperature of the particles that are coming from the Sun and forest fires in the USA. For determination of the degree of randomness for time series of input (temperature of protons) and output parameters (number of forest fires), the R/S analysis is conducted. The analysis of fractal dimension provides us an opportunity to compare self-similar processes in the influx of protons and the time series of forest fires flashes. Therefore we developed and conducted the sensitivity analysis of model based on hybrid neural networks ANFIS. As the calculations showed, only 16% of the real forest flashes cannot be predicted by the model.
PB  - Belgrade :  Vinča Institute of Nuclear Sciences
T2  - Thermal Science
T1  - Modelling of forest fires time evolution in the USA on the basis of long term variations and dynamics of the temperature of the solar wind protons
SP  - 437
EP  - 444
VL  - 19
IS  - 2
DO  - 10.2298/TSCI141103150R
UR  - https://hdl.handle.net/21.15107/rcub_dais_13498
ER  - 

@article{
author = "Radovanović, Milan M. and Vyklyuk, Yaroslav and Malinović-Milićević, Slavica B. and Jakovljević, Dejana M. and Pecelj, Milica R.",
year = "2015",
abstract = "The work examines the potential causative link between the flow of protons, i. e. temperature of the particles that are coming from the Sun and forest fires in the USA. For determination of the degree of randomness for time series of input (temperature of protons) and output parameters (number of forest fires), the R/S analysis is conducted. The analysis of fractal dimension provides us an opportunity to compare self-similar processes in the influx of protons and the time series of forest fires flashes. Therefore we developed and conducted the sensitivity analysis of model based on hybrid neural networks ANFIS. As the calculations showed, only 16% of the real forest flashes cannot be predicted by the model.",
publisher = "Belgrade :  Vinča Institute of Nuclear Sciences",
journal = "Thermal Science",
title = "Modelling of forest fires time evolution in the USA on the basis of long term variations and dynamics of the temperature of the solar wind protons",
pages = "437-444",
volume = "19",
number = "2",
doi = "10.2298/TSCI141103150R",
url = "https://hdl.handle.net/21.15107/rcub_dais_13498"
}

Radovanović, M. M., Vyklyuk, Y., Malinović-Milićević, S. B., Jakovljević, D. M.,& Pecelj, M. R.. (2015). Modelling of forest fires time evolution in the USA on the basis of long term variations and dynamics of the temperature of the solar wind protons. in Thermal Science
Belgrade :  Vinča Institute of Nuclear Sciences., 19(2), 437-444.
https://doi.org/10.2298/TSCI141103150R
https://hdl.handle.net/21.15107/rcub_dais_13498

Radovanović MM, Vyklyuk Y, Malinović-Milićević SB, Jakovljević DM, Pecelj MR. Modelling of forest fires time evolution in the USA on the basis of long term variations and dynamics of the temperature of the solar wind protons. in Thermal Science. 2015;19(2):437-444.
doi:10.2298/TSCI141103150R
https://hdl.handle.net/21.15107/rcub_dais_13498 .

Radovanović, Milan M., Vyklyuk, Yaroslav, Malinović-Milićević, Slavica B., Jakovljević, Dejana M., Pecelj, Milica R., "Modelling of forest fires time evolution in the USA on the basis of long term variations and dynamics of the temperature of the solar wind protons" in Thermal Science, 19, no. 2 (2015):437-444,
https://doi.org/10.2298/TSCI141103150R .,
https://hdl.handle.net/21.15107/rcub_dais_13498 .