A. Nechyporenko, Yevhen Hubarenko, Maryna Hubarenko
This paper analyses the topology of a regional power system distribution grid. This research improves the efficiency of the electric power system's operation by using upgrading the redesign (reengineering) methods of topological structures within distribution grids. The research object is an electric power system that consists of generation, transmission and distribution parts and requires reengineering. The subject of research is the re-engineering of the topology of the power system distribution grid. To achieve the research purpose, modifications of k-means algorithm as well as the small step algorithm based on the statistical analysis, clustering and minimum spanning tree development methods of Prim and Kraskal are used. The modifications described in this paper allow for optimization of the network based on user needs, properties of the operating grid elements, and other additional constraints for flexibility and generality. Given the varying parameters, this method provides the means to redesign parts of the distribution grid, keeping its certain elements safe from displacement, but also the means to redesign the whole distribution grid, including changes to the number and location of transformer substations and transmission lines. Conclusions. To solve the problem of determining the territorially close groups of consumers in the paper, it was proposed to use the k-means clustering algorithm. This algorithm allows us to divide consumer sets into clusters, so that coordinates of their centres will be recommended as locations of transformer substations. The modernization of the k-means algorithm was proposed by developing procedures for adding and combining clusters using different strategies for determining starting centroids. Based on this, a method for reengineering the topological structures of regional electric power systems in terms of the possibility of their fundamental restructuring was developed. The results of this research may be useful to various enterprises, organizations or institutes dealing with the elaboration or design of electric power system development on the corporate, regional or local level.
{"title":"Topological structure reengineering regional electric power systems","authors":"A. Nechyporenko, Yevhen Hubarenko, Maryna Hubarenko","doi":"10.32620/reks.2022.2.04","DOIUrl":"https://doi.org/10.32620/reks.2022.2.04","url":null,"abstract":"This paper analyses the topology of a regional power system distribution grid. This research improves the efficiency of the electric power system's operation by using upgrading the redesign (reengineering) methods of topological structures within distribution grids. The research object is an electric power system that consists of generation, transmission and distribution parts and requires reengineering. The subject of research is the re-engineering of the topology of the power system distribution grid. To achieve the research purpose, modifications of k-means algorithm as well as the small step algorithm based on the statistical analysis, clustering and minimum spanning tree development methods of Prim and Kraskal are used. The modifications described in this paper allow for optimization of the network based on user needs, properties of the operating grid elements, and other additional constraints for flexibility and generality. Given the varying parameters, this method provides the means to redesign parts of the distribution grid, keeping its certain elements safe from displacement, but also the means to redesign the whole distribution grid, including changes to the number and location of transformer substations and transmission lines. Conclusions. To solve the problem of determining the territorially close groups of consumers in the paper, it was proposed to use the k-means clustering algorithm. This algorithm allows us to divide consumer sets into clusters, so that coordinates of their centres will be recommended as locations of transformer substations. The modernization of the k-means algorithm was proposed by developing procedures for adding and combining clusters using different strategies for determining starting centroids. Based on this, a method for reengineering the topological structures of regional electric power systems in terms of the possibility of their fundamental restructuring was developed. The results of this research may be useful to various enterprises, organizations or institutes dealing with the elaboration or design of electric power system development on the corporate, regional or local level.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43081438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The subject of the research is the models of artificial intelligence (AI) quality. The current paper develops an AI quality model based on the definition and ordering of its characteristics. Objectives: to develop the principles and justify the sequence of analysis and development of AI quality models as ordered sets of characteristics; to offer models of AI quality for further use, first, the evaluation of individual characteristics and quality in general; to demonstrate the profiling of AI quality models for systems using artificial intelligence. The following results were obtained. The sequence of construction of AI quality models is offered. Based on the analysis of references, a list of AI characteristics was formed and their definitions were harmonized. The general model of AI quality is presented with a description of the step-by-step procedure for the realization of its hierarchical construction. A basic model of AI with abbreviated sets of characteristics is proposed due to its importance. Examples of profiling of quality models for two systems - monitoring of engineering communications and recognition of road signs are given. Conclusions. The study's main result is the development of a quality model for artificial intelligence, which is based on the analysis and harmonization of definitions and dependencies of quality characteristics specific to AI. The selection of characteristics and the construction of the quality model were carried out in such a way to exclude duplication, ensure the completeness of the presentation, as well as to determine the specific features of each characteristic. It is extremely difficult to create a model that would fully meet such requirements, so the presented options should be supplemented and improved considering the rapid development of technologies and applications of AI. The proposed quality models are open and can be supplemented and detailed according to the specific purpose and scope of AI.
{"title":"Базова модель нефункційних характеристик для оцінки якості штучного інтелекту","authors":"V. Kharchenko, H. Fesenko, O. Illiashenko","doi":"10.32620/reks.2022.2.11","DOIUrl":"https://doi.org/10.32620/reks.2022.2.11","url":null,"abstract":"The subject of the research is the models of artificial intelligence (AI) quality. The current paper develops an AI quality model based on the definition and ordering of its characteristics. Objectives: to develop the principles and justify the sequence of analysis and development of AI quality models as ordered sets of characteristics; to offer models of AI quality for further use, first, the evaluation of individual characteristics and quality in general; to demonstrate the profiling of AI quality models for systems using artificial intelligence. The following results were obtained. The sequence of construction of AI quality models is offered. Based on the analysis of references, a list of AI characteristics was formed and their definitions were harmonized. The general model of AI quality is presented with a description of the step-by-step procedure for the realization of its hierarchical construction. A basic model of AI with abbreviated sets of characteristics is proposed due to its importance. Examples of profiling of quality models for two systems - monitoring of engineering communications and recognition of road signs are given. Conclusions. The study's main result is the development of a quality model for artificial intelligence, which is based on the analysis and harmonization of definitions and dependencies of quality characteristics specific to AI. The selection of characteristics and the construction of the quality model were carried out in such a way to exclude duplication, ensure the completeness of the presentation, as well as to determine the specific features of each characteristic. It is extremely difficult to create a model that would fully meet such requirements, so the presented options should be supplemented and improved considering the rapid development of technologies and applications of AI. The proposed quality models are open and can be supplemented and detailed according to the specific purpose and scope of AI.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43642436","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
K. Dergachov, L. Krasnov, Vladislav Bilozerskyi, Anatolii Zymovin
The subject of research. There has been studying a new universal method of information protection in optical text recognition systems when transmitting confidential data over open communication channels. This work develops the concept of creating a modern, simple and reliable method for protecting information during its transmission over communication channels, to determine the objective criteria for the quality of its work, to create a set of algorithms for implementing the proposed method and software for conducting experimental studies. The current work puts on the concept of creation of a simple and reliable current method for protecting information when passing it through communication channels, also to define the objective criteria for assessing the tool operation quality and to exercise the dedicated programs, which implement the proposed methods and developed algorithms. Based on the results of these studies, it must evaluate the practice effectiveness of the proposed method in terms of both the transmitted data coding/decoding reliability and the secrecy of the fact of special information transmission. Results. It is described the universal concept of producing and use of the contemporary methods of information protection in optical text recognition systems in a confidential data transmission over open communication channels. The main criteria for these systems performance quality are determined. A new combined method for encrypting transmitted messages using QR-codes with subsequent masking of the fact of data transmission by various methods of LSB-steganography is proposed. To conduct experimental studies, a text recognition program based on Tesseract OCR software version 4.0 was developed. The program in Python uses the recent resources of the OpenCV library. The dedicated software technique contributed to assessing the efficiency of the algorithms, which realized the transmitted data encryption and therefore communication links privacy. There are examples of the system operation and results of the software testing in modes of messages encoding for subsequent hidden transmission. Conclusion. The case studies acknowledge the high efficiency of the proposed method of confidential data protection when transmitting them via open networks. The technique can be taken as a basis for developing software aimed at protecting information in OCR systems offered by various manufacturers.
{"title":"Development of tools for information protection of optical text recognition systems","authors":"K. Dergachov, L. Krasnov, Vladislav Bilozerskyi, Anatolii Zymovin","doi":"10.32620/reks.2022.2.13","DOIUrl":"https://doi.org/10.32620/reks.2022.2.13","url":null,"abstract":"The subject of research. There has been studying a new universal method of information protection in optical text recognition systems when transmitting confidential data over open communication channels. This work develops the concept of creating a modern, simple and reliable method for protecting information during its transmission over communication channels, to determine the objective criteria for the quality of its work, to create a set of algorithms for implementing the proposed method and software for conducting experimental studies. The current work puts on the concept of creation of a simple and reliable current method for protecting information when passing it through communication channels, also to define the objective criteria for assessing the tool operation quality and to exercise the dedicated programs, which implement the proposed methods and developed algorithms. Based on the results of these studies, it must evaluate the practice effectiveness of the proposed method in terms of both the transmitted data coding/decoding reliability and the secrecy of the fact of special information transmission. Results. It is described the universal concept of producing and use of the contemporary methods of information protection in optical text recognition systems in a confidential data transmission over open communication channels. The main criteria for these systems performance quality are determined. A new combined method for encrypting transmitted messages using QR-codes with subsequent masking of the fact of data transmission by various methods of LSB-steganography is proposed. To conduct experimental studies, a text recognition program based on Tesseract OCR software version 4.0 was developed. The program in Python uses the recent resources of the OpenCV library. The dedicated software technique contributed to assessing the efficiency of the algorithms, which realized the transmitted data encryption and therefore communication links privacy. There are examples of the system operation and results of the software testing in modes of messages encoding for subsequent hidden transmission. Conclusion. The case studies acknowledge the high efficiency of the proposed method of confidential data protection when transmitting them via open networks. The technique can be taken as a basis for developing software aimed at protecting information in OCR systems offered by various manufacturers.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42877973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The application of unidirectional synchronization of two coupled Chen systems is exhibited in this work. In spite of the high dependence on initial conditions, which means that two initially close phase trajectories with time become uncorrelated, it is possible to synchronize two dynamic systems to make them evolve identically. Data transmission using chaos requires mixing an information signal with a chaotic carrier. This procedure performs data encryption and spreads the spectrum of an information signal, which increases information security and reliability. Thus, the prospect of using devices with chaotic dynamics in modern telecommunication and telemetry applications is due to several factors, including high information capacity, various frequencies, and confidentiality of messages. The proposed scheme is considered to be used in a measuring transducer design that requires sensors to operate at a long distance from the rest of the scheme. We propose an application of a chaotic oscillator as a transceiver module for a quarts sensor transducer, which could be used in a telemetry application. The process of producing non-periodic but determined oscillations by the non-linear Chen system and signal transmission application, based on it, are the subject of the research. The complete synchronization of two unidirectionally connected Chen systems and its signal transmission application are considered. The goal is to develop a transceiver extension for the quartz measuring transducer scheme to ensure the stable operation of sensors at a long distance from the rest of the scheme. The result of the research: a chaos synchronization scheme was applied to transmit a frequency-modulated signal, obtained from a difference-frequency block of the quartz sensor transducer. Additionally, the mathematical model and numerical modeling of the Chen dynamical system has been done. The numerical solution of the system's differential equations was obtained using Matlab software. To study the change in the dynamic regime depending on the parameters of the model, the spectrum of Lyapunov exponents was calculated and bifurcation diagrams were constructed. The circuit design of the Chen oscillator was built using Multisim software, which uses the PSpice model to simulate electrical components. A model of an analog signal transmission system with chaotic mixing of a frequency output signal with a chaotic carrier has been proposed as an extension of the use of quartz transducers in measuring devices.
{"title":"Chen system-based chaotic transceiver for frequency output quartz transducers","authors":"Serhii Pidchenko, A. Taranchuk, Maksym Slobodian","doi":"10.32620/reks.2022.2.14","DOIUrl":"https://doi.org/10.32620/reks.2022.2.14","url":null,"abstract":"The application of unidirectional synchronization of two coupled Chen systems is exhibited in this work. In spite of the high dependence on initial conditions, which means that two initially close phase trajectories with time become uncorrelated, it is possible to synchronize two dynamic systems to make them evolve identically. Data transmission using chaos requires mixing an information signal with a chaotic carrier. This procedure performs data encryption and spreads the spectrum of an information signal, which increases information security and reliability. Thus, the prospect of using devices with chaotic dynamics in modern telecommunication and telemetry applications is due to several factors, including high information capacity, various frequencies, and confidentiality of messages. The proposed scheme is considered to be used in a measuring transducer design that requires sensors to operate at a long distance from the rest of the scheme. We propose an application of a chaotic oscillator as a transceiver module for a quarts sensor transducer, which could be used in a telemetry application. The process of producing non-periodic but determined oscillations by the non-linear Chen system and signal transmission application, based on it, are the subject of the research. The complete synchronization of two unidirectionally connected Chen systems and its signal transmission application are considered. The goal is to develop a transceiver extension for the quartz measuring transducer scheme to ensure the stable operation of sensors at a long distance from the rest of the scheme. The result of the research: a chaos synchronization scheme was applied to transmit a frequency-modulated signal, obtained from a difference-frequency block of the quartz sensor transducer. Additionally, the mathematical model and numerical modeling of the Chen dynamical system has been done. The numerical solution of the system's differential equations was obtained using Matlab software. To study the change in the dynamic regime depending on the parameters of the model, the spectrum of Lyapunov exponents was calculated and bifurcation diagrams were constructed. The circuit design of the Chen oscillator was built using Multisim software, which uses the PSpice model to simulate electrical components. A model of an analog signal transmission system with chaotic mixing of a frequency output signal with a chaotic carrier has been proposed as an extension of the use of quartz transducers in measuring devices.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42819912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Микола Васильович Нечипорук, Олег Євгенович Федорович, Віктор Васильович Попов, Максим Сергійович Романов
The problem of staffing for the implementation of projects to create innovative products of aerospace technology (ACT) is developed and solved. The relevance of the study is related to the modeling of specialist profiles to acquire the competencies necessary to complete the project on the new AST product creation. The study creates the set of competencies necessary for the implementation of the project to create AST products by training (retraining) specialists, considering the requirements of the industry professional standards (IPS) and the requirements competencies necessary for the new project. Much attention is paid to the architecture of the complex AST product, which is presented as a set of components. The basic product (BP) representation is done in the form of typical components which creation is associated with the necessary specialties and competencies that meet the requirements of the IPS. Specialties contain many competencies, which, in their turn, are divided into knowledge, skills, and abilities. The project to create the new AST product includes the product architecture, which may differ from the BP architecture and therefore requires the activities to develop the new components. To represent the component architecture, specialties, and competencies of the developers, the precedent base (PBBP) is used. The comparison of new product architectures and BP is carried out using both qualitative and quantitative assessments. For qualitative assessments, linguistic variables and the lexicographic ordering of variants are used. Quantitative evaluations are used to compare and evaluate the proximity of the technical characteristics of the components of the new product and BP. The assessment of the closeness of the specialties and competencies required in the new project and existing in the PBBP is carried out using qualitative assessments presented in the form of letters of the Latin alphabet. An optimization model was developed to minimize the difference in competencies based on scores. The optimization is carried out using integer (Boolean) linear programming. Admissible values of costs, time, and risks associated with the training (retraining) of specialists for the implementation of the project to create the new AST product are used as restrictions. An example of how to create the competencies for the implementation of the project to create helicopter blades from composite materials is given. The scientific novelty of the study is related to the development of a method to staff the implementation of innovative projects to create the new AST products using component and competence-based approaches. It is recommended to use the results of this study to staff the implementation of the project created the innovative AST product with a complex component architecture.
{"title":"Моделювання профілів спеціалістів для планування та виконання проєктів зі створення інноваційних виробів аерокосмічної техніки","authors":"Микола Васильович Нечипорук, Олег Євгенович Федорович, Віктор Васильович Попов, Максим Сергійович Романов","doi":"10.32620/reks.2022.1.02","DOIUrl":"https://doi.org/10.32620/reks.2022.1.02","url":null,"abstract":"The problem of staffing for the implementation of projects to create innovative products of aerospace technology (ACT) is developed and solved. The relevance of the study is related to the modeling of specialist profiles to acquire the competencies necessary to complete the project on the new AST product creation. The study creates the set of competencies necessary for the implementation of the project to create AST products by training (retraining) specialists, considering the requirements of the industry professional standards (IPS) and the requirements competencies necessary for the new project. Much attention is paid to the architecture of the complex AST product, which is presented as a set of components. The basic product (BP) representation is done in the form of typical components which creation is associated with the necessary specialties and competencies that meet the requirements of the IPS. Specialties contain many competencies, which, in their turn, are divided into knowledge, skills, and abilities. The project to create the new AST product includes the product architecture, which may differ from the BP architecture and therefore requires the activities to develop the new components. To represent the component architecture, specialties, and competencies of the developers, the precedent base (PBBP) is used. The comparison of new product architectures and BP is carried out using both qualitative and quantitative assessments. For qualitative assessments, linguistic variables and the lexicographic ordering of variants are used. Quantitative evaluations are used to compare and evaluate the proximity of the technical characteristics of the components of the new product and BP. The assessment of the closeness of the specialties and competencies required in the new project and existing in the PBBP is carried out using qualitative assessments presented in the form of letters of the Latin alphabet. An optimization model was developed to minimize the difference in competencies based on scores. The optimization is carried out using integer (Boolean) linear programming. Admissible values of costs, time, and risks associated with the training (retraining) of specialists for the implementation of the project to create the new AST product are used as restrictions. An example of how to create the competencies for the implementation of the project to create helicopter blades from composite materials is given. The scientific novelty of the study is related to the development of a method to staff the implementation of innovative projects to create the new AST products using component and competence-based approaches. It is recommended to use the results of this study to staff the implementation of the project created the innovative AST product with a complex component architecture.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49616514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
V. Moiseenko, Оleksandra Golovko, V. Butenko, Karyna Trubchaninova
The subject of the article is the processes of synthesis of a mathematical model of control objects functioning in computer information-control systems of critical purpose for the needs of high-speed railway transport. The main emphasis is on modeling the movement of a passenger train in the high-speed system of Ukrzaliznytsia. The aim is to study the process of regulating the speed of railway vehicles under conditions of uncertainty in the primary information of microprocessor information-control systems of railway transport. Tasks: determination of the criterion of the safety of railway vehicle auto control; obtain a mathematical model of train movement under conditions of uncertainty; check the adequacy of the model. The method used is the mathematical apparatus of discrete models. The following results have been obtained. The mathematical model of train movement developed in this work includes not only information on train position, reference point, direction, and speed of the vehicle but also a variable control indicator to reflect the process of railway traffic adequately. The study shows that, based on the synthesized model, it is possible to use the so-called fuzzy distance between adjacent trains. This approach improves the accuracy of determining the critical distance between trains, the time required to eliminate the risk of collision, the start time of braking, and braking time considering the angle of inclination of the track, as well as the distance of the braking distance. The necessity to determine the control indicator, its value for many points of time, while there is a reduction in speed for the safe movement of trains. Based on the proposed mathematical model, a computer simulation of the process was performed to determine the required time reserve for the train driver to respond to changes in the speed of the previous train, as well as speed ranges that require immediate emergency action. Conclusions. The scientific novelty of the obtained results is the development of a mathematical model of the behavior of mobile units in computer systems for critical use for the needs of railway transport in the presence of failures in the primary information from sensors that record motion parameters. The behavior of the control system at different values of train speed and changes in the value of the interval of the accompanying journey is studied. The theory of traction calculations in computer control systems for mobile units has been further developed. The obtained scientific results will be used in the development of an application program for many critical computer systems for railway.
{"title":"Modeling of vehicle movement in computer information-control systems","authors":"V. Moiseenko, Оleksandra Golovko, V. Butenko, Karyna Trubchaninova","doi":"10.32620/reks.2022.1.03","DOIUrl":"https://doi.org/10.32620/reks.2022.1.03","url":null,"abstract":"The subject of the article is the processes of synthesis of a mathematical model of control objects functioning in computer information-control systems of critical purpose for the needs of high-speed railway transport. The main emphasis is on modeling the movement of a passenger train in the high-speed system of Ukrzaliznytsia. The aim is to study the process of regulating the speed of railway vehicles under conditions of uncertainty in the primary information of microprocessor information-control systems of railway transport. Tasks: determination of the criterion of the safety of railway vehicle auto control; obtain a mathematical model of train movement under conditions of uncertainty; check the adequacy of the model. The method used is the mathematical apparatus of discrete models. The following results have been obtained. The mathematical model of train movement developed in this work includes not only information on train position, reference point, direction, and speed of the vehicle but also a variable control indicator to reflect the process of railway traffic adequately. The study shows that, based on the synthesized model, it is possible to use the so-called fuzzy distance between adjacent trains. This approach improves the accuracy of determining the critical distance between trains, the time required to eliminate the risk of collision, the start time of braking, and braking time considering the angle of inclination of the track, as well as the distance of the braking distance. The necessity to determine the control indicator, its value for many points of time, while there is a reduction in speed for the safe movement of trains. Based on the proposed mathematical model, a computer simulation of the process was performed to determine the required time reserve for the train driver to respond to changes in the speed of the previous train, as well as speed ranges that require immediate emergency action. Conclusions. The scientific novelty of the obtained results is the development of a mathematical model of the behavior of mobile units in computer systems for critical use for the needs of railway transport in the presence of failures in the primary information from sensors that record motion parameters. The behavior of the control system at different values of train speed and changes in the value of the interval of the accompanying journey is studied. The theory of traction calculations in computer control systems for mobile units has been further developed. The obtained scientific results will be used in the development of an application program for many critical computer systems for railway.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42408042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The problem of identifying the parameters of a linear object in the presence of non-Gaussian noise is considered. The identification algorithm is a gradient procedure for maximizing the functional, which is a correntropy. This functionality allows you to get estimates that have robust properties. In contrast to the commonly used Gaussian kernels, the centers of which are at zero and effective for distributions with zero mean, the paper considers a modification of the criterion suitable for distributions with nonzero mean. The modification is to use correntropy with a variable center The use of Gaussian kernels with a variable center will allow us to estimate unknown parameters under Gaussian and non-Gaussian noises with zero and non-zero mean distributions and provide an opportunity to develop new technologies for data analysis and processing. It is important to develop a robust identification algorithm based on correntropy with variable center. Their properties in the identification of stationary and non-stationary objects are the subject of research. The goal is to develop a robust identification algorithm that maximizes the criterion of correntropy with a variable center using center configuration procedures and kernel width and to study its convergence in stationary and non-stationary cases under non-Gaussian noise. Expressions for steady-state value of the estimation error are obtained, which depend on the type of noise distribution and the degree of non-stationarity of the estimated parameters The following tasks are solved: to investigate the convergence of the algorithm and determine the conditions for the stability of the established identification process. Methods of estimation theory (identification) and probability theory are used. The following results were obtained: 1) the developed algorithm provides robust estimates in the presence of noises having a distribution with zero and non-zero mean; 2) its convergence was studied in stationary and non-stationary cases under conditions of Gaussian and non-Gaussian noise; 3) simulation of the algorithm was carried out. 1) the developed algorithm consists in the development of a robust identification algorithm that maximizes the criterion of correntropy with a variable center; 2) its convergence in stationary and non-stationary cases in the conditions of Gaussian and non-Gaussian noises is investigated; 3) simulation of the algorithm is performed. Conclusions: The results of the current study will improve existing data processing technologies based on robust estimates and accelerate the development of new computing programs in real time.
{"title":"Adaptive identification under the maximum correntropy criterion with variable center","authors":"O. Rudenko, O. Bezsonov","doi":"10.32620/reks.2022.1.17","DOIUrl":"https://doi.org/10.32620/reks.2022.1.17","url":null,"abstract":"The problem of identifying the parameters of a linear object in the presence of non-Gaussian noise is considered. The identification algorithm is a gradient procedure for maximizing the functional, which is a correntropy. This functionality allows you to get estimates that have robust properties. In contrast to the commonly used Gaussian kernels, the centers of which are at zero and effective for distributions with zero mean, the paper considers a modification of the criterion suitable for distributions with nonzero mean. The modification is to use correntropy with a variable center The use of Gaussian kernels with a variable center will allow us to estimate unknown parameters under Gaussian and non-Gaussian noises with zero and non-zero mean distributions and provide an opportunity to develop new technologies for data analysis and processing. It is important to develop a robust identification algorithm based on correntropy with variable center. Their properties in the identification of stationary and non-stationary objects are the subject of research. The goal is to develop a robust identification algorithm that maximizes the criterion of correntropy with a variable center using center configuration procedures and kernel width and to study its convergence in stationary and non-stationary cases under non-Gaussian noise. Expressions for steady-state value of the estimation error are obtained, which depend on the type of noise distribution and the degree of non-stationarity of the estimated parameters The following tasks are solved: to investigate the convergence of the algorithm and determine the conditions for the stability of the established identification process. Methods of estimation theory (identification) and probability theory are used. The following results were obtained: 1) the developed algorithm provides robust estimates in the presence of noises having a distribution with zero and non-zero mean; 2) its convergence was studied in stationary and non-stationary cases under conditions of Gaussian and non-Gaussian noise; 3) simulation of the algorithm was carried out. 1) the developed algorithm consists in the development of a robust identification algorithm that maximizes the criterion of correntropy with a variable center; 2) its convergence in stationary and non-stationary cases in the conditions of Gaussian and non-Gaussian noises is investigated; 3) simulation of the algorithm is performed. Conclusions: The results of the current study will improve existing data processing technologies based on robust estimates and accelerate the development of new computing programs in real time.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"69960556","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Dotsenko, V. Kharchenko, O. Morozova, A. Rucinski, S. Dotsenko
From the analysis of the main theoretical provisions of heuristic self-organization systems and logical models, it follows that according to O. G. Ivakhnenko's systems of heuristic self-organization, the first task is to determine the factors content “that determine the essence of different images”. These are the images that characterize the objects of a particular subject area. After determining the composition and content of these images, the next problem is solved, namely, the problem of “generating the new successful heuristic”, which in content is a solution that leads to increased accuracy. Note that we are talking about improving the accuracy of solving the problem of data processing. It follows from the above mentioned that heuristic self-organization systems are data processing systems. This allows the multiplicity of heuristics. Heuristics in content correspond to the logical rules applied in heuristic self-organization systems. The main provisions of the heuristic self-organization system theory were developed by O. G. Ivakhnenko in the eighties of the last century, but they remain unnoticed to this day. At this time, the task is to explain why the neural network makes such a decision and not another. Based on this, the concept of “explainability of artificial intelligence” was introduced for artificial intelligence. It is the content of heuristics that forms the structure of the neural network in the form of logical rules and determines the logic of the decision made. It is established that the derivation rule, which is the basis for constructing artificial neural networks, is an abductive rule, which, unfortunately, does not meet the fourth heuristic and does not meet the definition of intelligence: intelligence is the ability to measure things. Unfortunately, none of the neural networks can measure things. From the analysis of the basic rules content of inference, it follows that the dialectical method of inference is general (generating) for the basic logical methods of inference. The difference lies in the composition and content of the middle member of the triangular relationship, namely, in the form of the element combination of the relationship: the transition from one concept to another. The explainability of artificial intelligence refers to the laws of the structure and activity of artificial neural networks. But modern theories of artificial neural networks ignore the existence of logical rules (heuristics), which were established by O. G. Ivakhnenko. After all, only knowing the rules based on which problems are solved, it is possible to check the correctness of the decision, but not by searching for such rules. The three hypotheses about the explainability of artificial intelligence and the theory of machine identification can be further defined as statements or theorems and strictly proved.
{"title":"Heuristic self-organization of knowledge representation and development: analysis in the context of explainable artificial intelligence","authors":"S. Dotsenko, V. Kharchenko, O. Morozova, A. Rucinski, S. Dotsenko","doi":"10.32620/reks.2022.1.04","DOIUrl":"https://doi.org/10.32620/reks.2022.1.04","url":null,"abstract":"From the analysis of the main theoretical provisions of heuristic self-organization systems and logical models, it follows that according to O. G. Ivakhnenko's systems of heuristic self-organization, the first task is to determine the factors content “that determine the essence of different images”. These are the images that characterize the objects of a particular subject area. After determining the composition and content of these images, the next problem is solved, namely, the problem of “generating the new successful heuristic”, which in content is a solution that leads to increased accuracy. Note that we are talking about improving the accuracy of solving the problem of data processing. It follows from the above mentioned that heuristic self-organization systems are data processing systems. This allows the multiplicity of heuristics. Heuristics in content correspond to the logical rules applied in heuristic self-organization systems. The main provisions of the heuristic self-organization system theory were developed by O. G. Ivakhnenko in the eighties of the last century, but they remain unnoticed to this day. At this time, the task is to explain why the neural network makes such a decision and not another. Based on this, the concept of “explainability of artificial intelligence” was introduced for artificial intelligence. It is the content of heuristics that forms the structure of the neural network in the form of logical rules and determines the logic of the decision made. It is established that the derivation rule, which is the basis for constructing artificial neural networks, is an abductive rule, which, unfortunately, does not meet the fourth heuristic and does not meet the definition of intelligence: intelligence is the ability to measure things. Unfortunately, none of the neural networks can measure things. From the analysis of the basic rules content of inference, it follows that the dialectical method of inference is general (generating) for the basic logical methods of inference. The difference lies in the composition and content of the middle member of the triangular relationship, namely, in the form of the element combination of the relationship: the transition from one concept to another. The explainability of artificial intelligence refers to the laws of the structure and activity of artificial neural networks. But modern theories of artificial neural networks ignore the existence of logical rules (heuristics), which were established by O. G. Ivakhnenko. After all, only knowing the rules based on which problems are solved, it is possible to check the correctness of the decision, but not by searching for such rules. The three hypotheses about the explainability of artificial intelligence and the theory of machine identification can be further defined as statements or theorems and strictly proved.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46095836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Vladyslav Yaloveha, A. Podorozhniak, Heorhii Kuchuk
In recent times, machine learning algorithms have shown great performance in solving problems in different fields of study, including the analysis of remote sensing images, computer vision, natural language processing, medical issues, etc. A well-prepared input dataset can have a huge impact on the result metrics. However, a correctly selected hyperparameter combined with neural network architecture could highly increase the final metrics. Therefore, the hyperparameters optimization problem becomes a key issue in a deep learning algorithm. The process of finding a suitable hyperparameter combination could be performed manually or automatically. Manual search is based on previous research and requires enormous human efforts. However, there are many automated hyperparameter optimization methods have been successfully applied in practice. The automated hyperparameter tuning techniques are divided into two groups: black-box optimization techniques (such as Grid Search, Random Search) and multi-fidelity optimization techniques (HyperBand, BOHB). The most recent and promising among all approaches is BOHB which, which combines both Bayesian optimization and bandit-based methods, outperforms classical approaches, and can run asynchronously with given GPU resources and time budget that plays a vital role in the hyperparameter optimization process. The previous study proposed a convolutional deep learning neural network for solving land cover classification problems in the EuroSAT dataset. It was found that adding spectral indexes NDVI, NDWI, and GNDVI with RGB channels increased the result accuracy (from 64.72% to 84.19%) and F1 (from 63.89 % to 84.05%) score. However, the convolutional neural network architecture and hyperparameter combination were selected manually. The research optimizes convolutional neural network architecture and finds suitable hyperparameter combinations applied to land cover classification problems using multispectral images. The obtained results must increase result performance compared with the previous study and given budget constraints.
{"title":"Convolutional neural network hyperparameter optimization applied to land cover classification","authors":"Vladyslav Yaloveha, A. Podorozhniak, Heorhii Kuchuk","doi":"10.32620/reks.2022.1.09","DOIUrl":"https://doi.org/10.32620/reks.2022.1.09","url":null,"abstract":"In recent times, machine learning algorithms have shown great performance in solving problems in different fields of study, including the analysis of remote sensing images, computer vision, natural language processing, medical issues, etc. A well-prepared input dataset can have a huge impact on the result metrics. However, a correctly selected hyperparameter combined with neural network architecture could highly increase the final metrics. Therefore, the hyperparameters optimization problem becomes a key issue in a deep learning algorithm. The process of finding a suitable hyperparameter combination could be performed manually or automatically. Manual search is based on previous research and requires enormous human efforts. However, there are many automated hyperparameter optimization methods have been successfully applied in practice. The automated hyperparameter tuning techniques are divided into two groups: black-box optimization techniques (such as Grid Search, Random Search) and multi-fidelity optimization techniques (HyperBand, BOHB). The most recent and promising among all approaches is BOHB which, which combines both Bayesian optimization and bandit-based methods, outperforms classical approaches, and can run asynchronously with given GPU resources and time budget that plays a vital role in the hyperparameter optimization process. The previous study proposed a convolutional deep learning neural network for solving land cover classification problems in the EuroSAT dataset. It was found that adding spectral indexes NDVI, NDWI, and GNDVI with RGB channels increased the result accuracy (from 64.72% to 84.19%) and F1 (from 63.89 % to 84.05%) score. However, the convolutional neural network architecture and hyperparameter combination were selected manually. The research optimizes convolutional neural network architecture and finds suitable hyperparameter combinations applied to land cover classification problems using multispectral images. The obtained results must increase result performance compared with the previous study and given budget constraints.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43254939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sergiy Kurennov, K. Barakhov, I. Taranenko, V. Stepanenko
A genetic algorithm for solving the problem of optimal beam material distribution along length at a given restriction on maximum sagging value is suggested. A review of literature sources is conducted and it was shown that existing solutions cover partial cases only in which the position of the point with maximum sagging was defined previously. In the paper presented I-section beam with constant proportions is considered, i.e., beam width, caps, and web thickness are proportional to beam height in the current cross-section. A statically determined beam is being considered. The load applied to a beam can be arbitrary, including cases of non-symmetrical loads and differently oriented ones. The position of point(s) at which beam sagging is maximum are unknown at the beginning of optimization and are found in the process solution. The problem is solved in the linear definition. Beam mass was assumed to be an optimization criterion. The method of finite differences is used for beam sagging finding, i.e., for the solution of the differential equation of the bending beam with a variable cross-section. Discretization allows transforming the problem of design into the problem of beam height determination at a system of reference points. At this stage, found values of beam height must satisfy restrictions on reference point displacements. The suggested technique allows controlling beam displacement quite flexibly because restrictions on point displacement are considered separately and do not depend on each other. The suggested objective function is the linear superposition of beam mass and the possible penalty in case of beam maximum sagging over exceeding predefined values. The application of a genetic algorithm allows getting sets of beam thicknesses those, which guaranty reaching the minimum of the objective function. The model problem is solved. It is shown that the suggested algorithm allows effectively solves problems of optimal design of beams with restrictions on the maximum sagging value. The suggested approach can be developed for strength restrictions, statically undetermined structures, etc.
{"title":"A genetic algorithm of optimal design of beam at restricted sagging","authors":"Sergiy Kurennov, K. Barakhov, I. Taranenko, V. Stepanenko","doi":"10.32620/reks.2022.1.06","DOIUrl":"https://doi.org/10.32620/reks.2022.1.06","url":null,"abstract":"A genetic algorithm for solving the problem of optimal beam material distribution along length at a given restriction on maximum sagging value is suggested. A review of literature sources is conducted and it was shown that existing solutions cover partial cases only in which the position of the point with maximum sagging was defined previously. In the paper presented I-section beam with constant proportions is considered, i.e., beam width, caps, and web thickness are proportional to beam height in the current cross-section. A statically determined beam is being considered. The load applied to a beam can be arbitrary, including cases of non-symmetrical loads and differently oriented ones. The position of point(s) at which beam sagging is maximum are unknown at the beginning of optimization and are found in the process solution. The problem is solved in the linear definition. Beam mass was assumed to be an optimization criterion. The method of finite differences is used for beam sagging finding, i.e., for the solution of the differential equation of the bending beam with a variable cross-section. Discretization allows transforming the problem of design into the problem of beam height determination at a system of reference points. At this stage, found values of beam height must satisfy restrictions on reference point displacements. The suggested technique allows controlling beam displacement quite flexibly because restrictions on point displacement are considered separately and do not depend on each other. The suggested objective function is the linear superposition of beam mass and the possible penalty in case of beam maximum sagging over exceeding predefined values. The application of a genetic algorithm allows getting sets of beam thicknesses those, which guaranty reaching the minimum of the objective function. The model problem is solved. It is shown that the suggested algorithm allows effectively solves problems of optimal design of beams with restrictions on the maximum sagging value. The suggested approach can be developed for strength restrictions, statically undetermined structures, etc.","PeriodicalId":36122,"journal":{"name":"Radioelectronic and Computer Systems","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48678706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：