The use of hydrocracked and straight-run vacuum residues in the production of road pavement bitum... more The use of hydrocracked and straight-run vacuum residues in the production of road pavement bitumen requires a good understanding of how the viscosity and softening point can be modeled and controlled. Scientific reports on modeling of these rheological properties for hydrocracked and straight-run vacuum residues are scarce. For that reason, 30 straight-run vacuum residues and 33 hydrocracked vacuum residues obtained in a conversion range of 55–93% were investigated, and the characterization data were employed for modeling purposes. An intercriteria analysis was applied to investigate the statistically meaningful relations between the studied vacuum residue properties. It revealed that the straight-run and hydrocracked vacuum residues were completely different, and therefore their viscosity and softening point should be separately modeled. Through the use of nonlinear regression by applying CAS Maple and NLPSolve with the modified Newton iterative method and the vacuum residue bulk ...
Recently, active developments of projects related to alternative energy sources can be observed, ... more Recently, active developments of projects related to alternative energy sources can be observed, including revival of long known effects like the effect of Seebeck. The present paper presents studies on three thermoelectric modules working in a generator mode. On the basis of the results obtained, a neural network for prediction of the parameters of the thermoelectric batteries made from such modules was synthesized.
A chemometric approach using artificial neural network for clusterization of biodiesels was devel... more A chemometric approach using artificial neural network for clusterization of biodiesels was developed. It is based on artificial ART2 neural network. Gas chromatography (GC) and Gas Chromatographymass spectrometry (GC-MS) were used for quantitative and qualitative analysis of biodiesels, produced from different feedstocks, and FAME (fatty acid methyl esters) profiles were determined. Totally 96 analytical results for 7 different classes of biofuel plants: sunflower, rapeseed, corn, soybean, palm, peanut, "unknown" were used as objects. The analysis of biodiesels showed the content of five major FAME (C16:0, C18:0, C18:1, C18:2, C18:3) and those components were used like inputs in the model. After training with 6 samples, for which the origin was known, ANN was verified and tested with ninety "unknown" samples. The present research demonstrated the successful application of neural network for recognition of biodiesels according to their feedstock which give information upon their properties and handling.
A chemometric approach using artificial neural network for classification of biodiesels was devel... more A chemometric approach using artificial neural network for classification of biodiesels was developed. It is based on artificial neural network in its classic form Multilayer Perceptron. Gas chromatography (GC) and Gas Chromatography-mass spectrometry (GC-MS) were used for quantitative and qualitative analysis of biodiesels, produced from different feedstocks, and FAME (fatty acid methyl esters) profiles were determined. Totally 93 analytical results for 7 different classes of biofuel plants: sunflower, rapeseed, corn, soybean, palm, peanut, "unknown" were used as objects. The analysis of biodiesels showed content of five major FAME (C16:0, C18:0, C18:1, C18:2, C18:3) and those components were used like inputs in the model. After training with 85 samples, for which the origin was known, ANN was tested with eight "unknown" samples. The "unknown" samples were properly recognized with an error between 1 and 4 %. The present research demonstrated the successful application of neural network for recognition of biodiesels according to their feedstock which give information upon their properties and handling.
In this paper a generalized net model of the Neocognitron neural network is presented. A Network ... more In this paper a generalized net model of the Neocognitron neural network is presented. A Network Neocognitron is a self-organizing network with the ability to recognize patterns based on the difference of their form. A neocognitron is able to correctly identify an image, even if there is a violation or movement into position. Self-organization in the neocognitron is also realized uncontrollably - training for self-organizing neocognitron takes only a collection of recurring patterns in the recognizable image and does not need the information for categories that include templates. The output producing process is presented by a Generalized net model.
This article is an open access article distributed under the terms and conditions of the Creative... more This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY
In this investigation the level of burnout among the medical employees was analyzed. Тhe InterCri... more In this investigation the level of burnout among the medical employees was analyzed. Тhe InterCriteria Analysis (ICA) approach is used to find the dependences between different parameters characterizing the 139 medical employees from 6 medical centers. The aim is to analyze the correlations between the health indicators, by surveying with a developed questionnaire. The obtained data from the InterCriteria Analysis were clustered using an adaptive neural network.
In this paper, a fractional-order Cohen–Grossberg-type neural network with Caputo fractional deri... more In this paper, a fractional-order Cohen–Grossberg-type neural network with Caputo fractional derivatives is investigated. The notion of almost periodicity is adapted to the impulsive generalization of the model. General types of impulsive perturbations not necessarily at fixed moments are considered. Criteria for the existence and uniqueness of almost periodic waves are proposed. Furthermore, the global perfect Mittag–Leffler stability notion for the almost periodic solution is defined and studied. In addition, a robust global perfect Mittag–Leffler stability analysis is proposed. Lyapunov-type functions and fractional inequalities are applied in the proof. Since the type of Cohen–Grossberg neural networks generalizes several basic neural network models, this research contributes to the development of the investigations on numerous fractional neural network models.
2020 IEEE 10th International Conference on Intelligent Systems (IS)
In a series of papers, Generalized Nets (GNs) are constructed representing the functioning and th... more In a series of papers, Generalized Nets (GNs) are constructed representing the functioning and the results of the work of different types of Neural Networks (NNs). In the present research, we show that the functioning and the results of the work of a given NN can be represented by a GN from a very simple type, but the so constructed GN-model can be used for extension of the concept of NNs.
In the paper, a method for evaluation of fingerprint equivalence obtained in a fingerprint recogn... more In the paper, a method for evaluation of fingerprint equivalence obtained in a fingerprint recognition system is proposed. For the assessment of the equivalence of the respective assessment units, the theory of intuitionistic fuzzy sets is used. The obtained intuitionistic fuzzy estimations reflect on the recognition of the system. We also consider a degree of uncertainty when the information is not enough. In this case we use threshold values for the minimum and maximum of the degree of membership and non-membership. For the description of the entire process, we use generalized nets model.
Advances in Intelligent Systems and Computing, 2020
Generalized Nets (GNs) are constructed in a series of papers, representing the functioning and th... more Generalized Nets (GNs) are constructed in a series of papers, representing the functioning and the results of the work of different types of Neural Networks (NNs). In the present research, we show the functioning and the results of the structure of a Convolutional Neural Networks.
Neural networks are a tool that can be used for the modelling of many systems and process behavio... more Neural networks are a tool that can be used for the modelling of many systems and process behavior. The artificial neural networks can “understand” the information from health care processes. For the estimations between these two concepts we use intuitionistic fuzzy sets. Here, for the learning process of the neural networks, we will use 60 heavy oils that have been characterized for their distillation characteristics by ASTM D-5236 and ASTM D-1160 in the Research laboratory of LUKOIL Neftochim Burgas. The aim is to recognize the type of crude oil based on six of their properties.
In the paper application of the InterCriteria analysis approach to real dataset with instances of... more In the paper application of the InterCriteria analysis approach to real dataset with instances of hourly averaged responses from an array of 5 metal oxide chemical sensors embedded in an air quality chemical multisensor device [29, 30] is represented. The InterCriteria analysis is a new method that can be used for multicriteria decision making. The aim is to analyze the correlations between 12 indicators representing the recordings of on field deployed air quality chemical sensor devices responses.
Three H-Oil gas oils, heavy atmospheric gas oil (HAGO), light vacuum gas oil (LVGO), heavy vacuum... more Three H-Oil gas oils, heavy atmospheric gas oil (HAGO), light vacuum gas oil (LVGO), heavy vacuum gas oil (HVGO), and two their blends with hydrotreated straight run vacuum gas oils (HTSRVGOs) were cracked on two high unit cell size (UCS) lower porosity commercial catalysts and two low UCS higher porosity commercial catalysts. The cracking experiments were performed in an advanced cracking evaluation fluid catalytic cracking (FCC) laboratory unit at 527°C, 30 s catalyst time on stream, and catalyst-to-oil (CTO) variation between 3.5 and 7.5 wt/wt The two high UCS lower porosity catalysts were more active and more coke selective. However, the difference between conversion of the more active high UCS lower porosity and low UCS higher porosity catalysts at 7.5 wt/wt CTO decreased in the order 10% (HAGO) > 9% (LVGO) > 6% (HVGO) > 4% (80% HTSRVGO/20% H-Oil VGO). Therefore, the catalyst performance is feedstock-dependent. The four studied catalysts along with a blend of one of them with 2% ZSM-5 were examined in a commercially revamped UOP FCC VSS unit. The lower UCS higher porosity catalysts exhibited operation at a higher CTO ratio achieving a similar conversion level with more active higher UCS lower porosity catalysts. However, the higher UCS lower porosity catalysts made 0.67% Δ coke that was higher than the maximum acceptable limit of 0.64% for this particular commercial FCC unit (FCCU), which required excluding the HVGO from the FCC feed blend. The catalyst system containing ZSM-5 increased the LPG yield but did not have an impact on gasoline octane. It was found that the predominant factor that controls refinery profitability related to the FCCU performance is the FCC slurry oil (bottoms) yield.
The exactitude of petroleum fluid molecular weight correlations affects significantly the precisi... more The exactitude of petroleum fluid molecular weight correlations affects significantly the precision of petroleum engineering calculations and can make process design and trouble-shooting inaccurate. Some of the methods in the literature to predict petroleum fluid molecular weight are used in commercial software process simulators. According to statements made in the literature, the correlations of Lee–Kesler and Twu are the most used in petroleum engineering, and the other methods do not exhibit any significant advantages over the Lee–Kesler and Twu correlations. In order to verify which of the proposed in the literature correlations are the most appropriate for petroleum fluids with molecular weight variation between 70 and 1685 g/mol, 430 data points for boiling point, specific gravity, and molecular weight of petroleum fluids and individual hydrocarbons were extracted from 17 literature sources. Besides the existing correlations in the literature, two different techniques, nonlin...
Uncertainty and Imprecision in Decision Making and Decision Support: Cross-Fertilization, New Models and Applications, 2017
The mathematical model for predicting the forest dynamics from [16] is extended with intuitionist... more The mathematical model for predicting the forest dynamics from [16] is extended with intuitionistic fuzzy estimations for the rules for Game Method for Modelling.
The use of hydrocracked and straight-run vacuum residues in the production of road pavement bitum... more The use of hydrocracked and straight-run vacuum residues in the production of road pavement bitumen requires a good understanding of how the viscosity and softening point can be modeled and controlled. Scientific reports on modeling of these rheological properties for hydrocracked and straight-run vacuum residues are scarce. For that reason, 30 straight-run vacuum residues and 33 hydrocracked vacuum residues obtained in a conversion range of 55–93% were investigated, and the characterization data were employed for modeling purposes. An intercriteria analysis was applied to investigate the statistically meaningful relations between the studied vacuum residue properties. It revealed that the straight-run and hydrocracked vacuum residues were completely different, and therefore their viscosity and softening point should be separately modeled. Through the use of nonlinear regression by applying CAS Maple and NLPSolve with the modified Newton iterative method and the vacuum residue bulk ...
Recently, active developments of projects related to alternative energy sources can be observed, ... more Recently, active developments of projects related to alternative energy sources can be observed, including revival of long known effects like the effect of Seebeck. The present paper presents studies on three thermoelectric modules working in a generator mode. On the basis of the results obtained, a neural network for prediction of the parameters of the thermoelectric batteries made from such modules was synthesized.
A chemometric approach using artificial neural network for clusterization of biodiesels was devel... more A chemometric approach using artificial neural network for clusterization of biodiesels was developed. It is based on artificial ART2 neural network. Gas chromatography (GC) and Gas Chromatographymass spectrometry (GC-MS) were used for quantitative and qualitative analysis of biodiesels, produced from different feedstocks, and FAME (fatty acid methyl esters) profiles were determined. Totally 96 analytical results for 7 different classes of biofuel plants: sunflower, rapeseed, corn, soybean, palm, peanut, "unknown" were used as objects. The analysis of biodiesels showed the content of five major FAME (C16:0, C18:0, C18:1, C18:2, C18:3) and those components were used like inputs in the model. After training with 6 samples, for which the origin was known, ANN was verified and tested with ninety "unknown" samples. The present research demonstrated the successful application of neural network for recognition of biodiesels according to their feedstock which give information upon their properties and handling.
A chemometric approach using artificial neural network for classification of biodiesels was devel... more A chemometric approach using artificial neural network for classification of biodiesels was developed. It is based on artificial neural network in its classic form Multilayer Perceptron. Gas chromatography (GC) and Gas Chromatography-mass spectrometry (GC-MS) were used for quantitative and qualitative analysis of biodiesels, produced from different feedstocks, and FAME (fatty acid methyl esters) profiles were determined. Totally 93 analytical results for 7 different classes of biofuel plants: sunflower, rapeseed, corn, soybean, palm, peanut, "unknown" were used as objects. The analysis of biodiesels showed content of five major FAME (C16:0, C18:0, C18:1, C18:2, C18:3) and those components were used like inputs in the model. After training with 85 samples, for which the origin was known, ANN was tested with eight "unknown" samples. The "unknown" samples were properly recognized with an error between 1 and 4 %. The present research demonstrated the successful application of neural network for recognition of biodiesels according to their feedstock which give information upon their properties and handling.
In this paper a generalized net model of the Neocognitron neural network is presented. A Network ... more In this paper a generalized net model of the Neocognitron neural network is presented. A Network Neocognitron is a self-organizing network with the ability to recognize patterns based on the difference of their form. A neocognitron is able to correctly identify an image, even if there is a violation or movement into position. Self-organization in the neocognitron is also realized uncontrollably - training for self-organizing neocognitron takes only a collection of recurring patterns in the recognizable image and does not need the information for categories that include templates. The output producing process is presented by a Generalized net model.
This article is an open access article distributed under the terms and conditions of the Creative... more This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY
In this investigation the level of burnout among the medical employees was analyzed. Тhe InterCri... more In this investigation the level of burnout among the medical employees was analyzed. Тhe InterCriteria Analysis (ICA) approach is used to find the dependences between different parameters characterizing the 139 medical employees from 6 medical centers. The aim is to analyze the correlations between the health indicators, by surveying with a developed questionnaire. The obtained data from the InterCriteria Analysis were clustered using an adaptive neural network.
In this paper, a fractional-order Cohen–Grossberg-type neural network with Caputo fractional deri... more In this paper, a fractional-order Cohen–Grossberg-type neural network with Caputo fractional derivatives is investigated. The notion of almost periodicity is adapted to the impulsive generalization of the model. General types of impulsive perturbations not necessarily at fixed moments are considered. Criteria for the existence and uniqueness of almost periodic waves are proposed. Furthermore, the global perfect Mittag–Leffler stability notion for the almost periodic solution is defined and studied. In addition, a robust global perfect Mittag–Leffler stability analysis is proposed. Lyapunov-type functions and fractional inequalities are applied in the proof. Since the type of Cohen–Grossberg neural networks generalizes several basic neural network models, this research contributes to the development of the investigations on numerous fractional neural network models.
2020 IEEE 10th International Conference on Intelligent Systems (IS)
In a series of papers, Generalized Nets (GNs) are constructed representing the functioning and th... more In a series of papers, Generalized Nets (GNs) are constructed representing the functioning and the results of the work of different types of Neural Networks (NNs). In the present research, we show that the functioning and the results of the work of a given NN can be represented by a GN from a very simple type, but the so constructed GN-model can be used for extension of the concept of NNs.
In the paper, a method for evaluation of fingerprint equivalence obtained in a fingerprint recogn... more In the paper, a method for evaluation of fingerprint equivalence obtained in a fingerprint recognition system is proposed. For the assessment of the equivalence of the respective assessment units, the theory of intuitionistic fuzzy sets is used. The obtained intuitionistic fuzzy estimations reflect on the recognition of the system. We also consider a degree of uncertainty when the information is not enough. In this case we use threshold values for the minimum and maximum of the degree of membership and non-membership. For the description of the entire process, we use generalized nets model.
Advances in Intelligent Systems and Computing, 2020
Generalized Nets (GNs) are constructed in a series of papers, representing the functioning and th... more Generalized Nets (GNs) are constructed in a series of papers, representing the functioning and the results of the work of different types of Neural Networks (NNs). In the present research, we show the functioning and the results of the structure of a Convolutional Neural Networks.
Neural networks are a tool that can be used for the modelling of many systems and process behavio... more Neural networks are a tool that can be used for the modelling of many systems and process behavior. The artificial neural networks can “understand” the information from health care processes. For the estimations between these two concepts we use intuitionistic fuzzy sets. Here, for the learning process of the neural networks, we will use 60 heavy oils that have been characterized for their distillation characteristics by ASTM D-5236 and ASTM D-1160 in the Research laboratory of LUKOIL Neftochim Burgas. The aim is to recognize the type of crude oil based on six of their properties.
In the paper application of the InterCriteria analysis approach to real dataset with instances of... more In the paper application of the InterCriteria analysis approach to real dataset with instances of hourly averaged responses from an array of 5 metal oxide chemical sensors embedded in an air quality chemical multisensor device [29, 30] is represented. The InterCriteria analysis is a new method that can be used for multicriteria decision making. The aim is to analyze the correlations between 12 indicators representing the recordings of on field deployed air quality chemical sensor devices responses.
Three H-Oil gas oils, heavy atmospheric gas oil (HAGO), light vacuum gas oil (LVGO), heavy vacuum... more Three H-Oil gas oils, heavy atmospheric gas oil (HAGO), light vacuum gas oil (LVGO), heavy vacuum gas oil (HVGO), and two their blends with hydrotreated straight run vacuum gas oils (HTSRVGOs) were cracked on two high unit cell size (UCS) lower porosity commercial catalysts and two low UCS higher porosity commercial catalysts. The cracking experiments were performed in an advanced cracking evaluation fluid catalytic cracking (FCC) laboratory unit at 527°C, 30 s catalyst time on stream, and catalyst-to-oil (CTO) variation between 3.5 and 7.5 wt/wt The two high UCS lower porosity catalysts were more active and more coke selective. However, the difference between conversion of the more active high UCS lower porosity and low UCS higher porosity catalysts at 7.5 wt/wt CTO decreased in the order 10% (HAGO) > 9% (LVGO) > 6% (HVGO) > 4% (80% HTSRVGO/20% H-Oil VGO). Therefore, the catalyst performance is feedstock-dependent. The four studied catalysts along with a blend of one of them with 2% ZSM-5 were examined in a commercially revamped UOP FCC VSS unit. The lower UCS higher porosity catalysts exhibited operation at a higher CTO ratio achieving a similar conversion level with more active higher UCS lower porosity catalysts. However, the higher UCS lower porosity catalysts made 0.67% Δ coke that was higher than the maximum acceptable limit of 0.64% for this particular commercial FCC unit (FCCU), which required excluding the HVGO from the FCC feed blend. The catalyst system containing ZSM-5 increased the LPG yield but did not have an impact on gasoline octane. It was found that the predominant factor that controls refinery profitability related to the FCCU performance is the FCC slurry oil (bottoms) yield.
The exactitude of petroleum fluid molecular weight correlations affects significantly the precisi... more The exactitude of petroleum fluid molecular weight correlations affects significantly the precision of petroleum engineering calculations and can make process design and trouble-shooting inaccurate. Some of the methods in the literature to predict petroleum fluid molecular weight are used in commercial software process simulators. According to statements made in the literature, the correlations of Lee–Kesler and Twu are the most used in petroleum engineering, and the other methods do not exhibit any significant advantages over the Lee–Kesler and Twu correlations. In order to verify which of the proposed in the literature correlations are the most appropriate for petroleum fluids with molecular weight variation between 70 and 1685 g/mol, 430 data points for boiling point, specific gravity, and molecular weight of petroleum fluids and individual hydrocarbons were extracted from 17 literature sources. Besides the existing correlations in the literature, two different techniques, nonlin...
Uncertainty and Imprecision in Decision Making and Decision Support: Cross-Fertilization, New Models and Applications, 2017
The mathematical model for predicting the forest dynamics from [16] is extended with intuitionist... more The mathematical model for predicting the forest dynamics from [16] is extended with intuitionistic fuzzy estimations for the rules for Game Method for Modelling.
Uploads
Papers by Sotir Sotirov