Synthesis and Parameterization of Gas Sensor Models

Purpose of research: search and analysis of existing models of gas-sensitive sensors. Development of mathematical models of gas-sensitive sensors of various types (semiconductor, thermocatalytic, optical, electrochemical) for their subsequent use in the training of artificial neural networks (INS). Investigation of main physicochemical patterns underlying the principles of sensor operation, consideration of the influence of environmental factors and cross-sensitivity on the sensor output signal. Comparison of simulation results with actual characteristics produced by the sensor industry. The concept of creating mathematical models is described. Their parameterization, research and assessment of adequacy are carried out.Methods. Numerical methods, computer modeling methods, electrical circuit theory, the theory of chemosorption and heterogeneous catalysis, the Freundlich and Langmuir equations, the Buger-Lambert-Behr law, the foundations of electrochemistry were used in creating mathematical models. Standard deviation (MSE) and relative error were calculated to assess the adequacy of the models.Results. The concept of creating mathematical models of sensors based on physicochemical patterns is described. This concept allows the process of data generation for training artificial neural networks used in multi-component gas analyzers for the purpose of joint information processing to be automated. Models of semiconductor, thermocatalytic, optical and electrochemical sensors were obtained and upgraded, considering the influence of additional factors on the sensor signal. Parameterization and assessment of adequacy and extrapolation properties of models by graphical dependencies presented in technical documentation of sensors were carried out. Errors (relative and RMS) of discrepancy of real data and results of simulation of gas-sensitive sensors by basic parameters are determined. The standard error of reproduction of the main characteristics of the sensors did not exceed 0.5%.Conclusion. Multivariable mathematical models of gas-sensitive sensors are synthesized, considering the influence of main gas and external factors (pressure, temperature, humidity, cross-sensitivity) on the output signal and allowing to generate training data for sensors of various types.

Applying Multitask Deep Learning to Emotion Recognition in Speech

Purpose of research. Emotions play one of the key roles in the regulation of human behaviour. Solving the problem of automatic recognition of emotions makes it possible to increase the effectiveness of operation of a whole range of digital systems such as security systems, human-machine interfaces, e-commerce systems, etc. At the same time, the low efficiency of modern approaches to recognizing emotions in speech can be noted. This work studies automatic recognition of emotions in speech applying machine learning methods.Methods. The article describes and tests an approach to automatic recognition of emotions in speech based on multitask learning of deep convolution neural networks of AlexNet and VGG architectures using automatic selection of the weight coefficients for each task when calculating the final loss value during learning. All the models were trained on a sample of the IEMOCAP dataset with four emotional categories of ‘anger’, ‘happiness’, ‘neutral emotion’, ‘sadness’. The log-mel spectrograms of statements processed by a specialized algorithm are used as input data.Results. The considered models were tested on the basis of numerical metrics: the share of correctly recognized instances, accuracy, completeness, f-measure. For all of the above metrics, an improvement in the quality of emotion recognition by the proposed model was obtained in comparison with the two basic single-task models as well as with known solutions. This result is achieved through the use of automatic weighting of the values of the loss functions from individual tasks when forming the final value of the error in the learning process.Conclusion. The resulting improvement in the quality of emotion recognition in comparison with the known solutions confirms the feasibility of applying multitask learning to increase the accuracy of emotion recognition models. The developed approach makes it possible to achieve a uniform and simultaneous reduction of errors of individual tasks, and is used in the field of emotions recognition in speech for the first time.

Recognition of character information for automation of production processes

Purpose of research. Nowadays optical character recognition systems have a high level of dependence on the specific type of marking that is to be recognized, and therefore, the creation of a universal solution is an important and difficult task. The paper considers the issue of creating a system for recognizing symbolic information that can be used at various stages of production to automate processes in control systems, in particular, to analyze the labeling of circuit breakers.Methods. Binarization, filtering, and boundary detection are digital image processing techniques. Line search method, baseline search method, word splitting algorithms, image enhancement methods by segmentation, damaged characters recognition method, an algorithm for increasing the final recognition quality are character recognition methods.Results. The analysis of algorithms used for preprocessing and subsequent recognition of images containing marking of circuit breakers is carried out. The mathematical model of image processing for subsequent recognition has been created. We have described methods used to define marking symbols. Illustrative examples of the operation of the algorithms on which the system is built are given. The obtained solution was tested. The ways of system development are described here, they can lead to improved results, for particular use cases.Conclusion. It is proposed a solution that recognizes symbolic information on the labeling of circuit breakers, which can be the basis for the development and description of systems serving for the automation of production, by transferring information read from the product during the production process. This system, by its example, describes the components of character recognition systems, and for direct use, it needs to be refined in accordance with the technical requirements and the specifics of the conditions in which it will be used.

Automated Leadership Selection Educational Projects Based on Genetic Algorithms

Purpose or research. The purpose of the study is to conduct a research of the possibility to use automated methods of finding a solution for the task of forming a selection of candidates in solving a wide range of business problems with the possibility of the influence of such factors as: requirements for the quality of business processes, restrictions on the competence of candidates, the employment of candidates in other processes of the company, planned urgency of fulfilling tasks, the amount of tasks in the pool and expected tasks, tasks characteristics, private policy of the company when selecting candidates, company's policy on risks.Methods. A typology of innovative business tasks is given by the subject area, the range of target results, the duration of execution, etc. External and internal factors are considered to ensure the effective operation of the project team and the success of the project. There are offered sets of attributive characteristics for assessment of projects quality and potential projects. The structure of the algorithm for solving the problem of creating a selection of candidates for solving a set of problems is considered. Prerequisites for using principles of genetic programming in solving the problem under consideration are given. Search algorithm implementation parameters, criteria and constraints are defined. Algorithm was implemented as well as modeling in the Jupyter Lab v2 environment. The results are described in the text. A relative analysis of practical effectiveness of the algorithm was carried out depending on the modeling parameters to justify the selection of their values.Results. The task of creating a selection of candidates for the implementation of projects pool considering a number of factors was described during the study. An approach to solving the problem based on a genetic heuristic search algorithm has been developed. A numerical simulation was performed in Jupyter Lab v2. Simulation results were analyzed, and algorithm parameters were selected.Conclusion. The proposed approach allows not only automatize the selection of managers based on the accumulated data history, but also to adjust the established process to change the vector of organization development. The interaction of education and informatics (information technology) can enrich and expand the field of both sciences in the field of recruitment of innovative project teams. Their object analysis, supplemented by genetic programming capabilities, together allows you to achieve specified qualities of the head of innovative project teams that help maximize business benefits while minimizing material costs. As a result of computational experiments using mathematical apparatus and genetic algorithm technologies, it is necessary to emphasize the possibility of extrapolating such approaches to any level of implementation of innovative projects.

PLC-based lumber humidity measurement method

Purpose or research. The paper considers the solution of the scientific and practical problem of measuring the humidity content of lumber on the basis of a programmable logic controller (PLC), which makes it possible to implement control systems for drying lumber on inexpensive controllers of wide application with the advantages of modern SCADA systems.Methods. To achieve this goal, modern methods for assessing the humidity content of lumber by electrical resistance have been investigated. We have provided a review and analysis of the known dependences of the electrical resistance of lumber on moisture. It is noted the need to measure resistance in the range of 2.5 KΩ to 25 MΩ. The schematic solution for matching a high-resistance measurement bridge with an additional voltage source with a PLC is described here. We have proposed a method for calculating the matching scheme. Expressions are obtained for calculating humidity according to the data of an analog-to-digital converter. The procedure for setting up the measurement channel according to the readings of two reference measurements is presented.Results. The estimation of the measurement error of humidity and resistance is carried out. It is shown that a resistance measurement error of up to 20% leads to a humidity measurement error of up to 1%, provided that the parameters are correctly configured. To set it up, it is enough to take two reference measurements to complete the procedure for setting the humidity measurement parameters.Conclusion. In the process of designing modern systems for drying lumber, you can use the algorithms and programs of the technological process available and well presented in the literature for implementation on the PLC. The article offers a solution to the main problem of measuring the humidity content of lumber, which is necessary for these algorithms and programs. The material can be useful for researching technological stages of the drying process and developing industrial control systems based on inexpensive controllers.

Experimental Research of Physical and Mechanical Characteristics of Reclamation Pipelines Made of Composite Materials

Purpose of research. Improving irrigation methods and techniques is the most important task of irrigated agriculture. In our country and abroad, a significant part of the area is watered by surface irrigation along furrows, strips and checks, which is most acceptable in the case of a sharp increase in the cost of energy carriers. More and more attention is being paid to the introduction of mechanical aids and automation of surface irrigation, as well as the development of various effective ways to distribute water across the field and reduce its losses. This is successfully achieved by using flexible lay­flat pipelines for irrigation. To increase irrigation productivity, improve quality and reliability, it is advisable to use structures with a reinforced seam or seamless lay-flat pipelines. The goal and objectives of the paper are the following ones: improving surface irrigation methods through the use of flexible reclamation pipelines that increase irrigation productivity and ensure efficient water distribution across the field; obtaining experimental characteristics of pipelines.Methods. According to the technical requirements and operating conditions, various physical and mechanical tests of fabric frames, polymer film coatings applied to the frame and pipelines with these coatings have been carried out. The experiments have been performed according to standard methods, as well as on a developed special installation.Results. The undertaken experiments confirmed the compliance of lay-flat pipelines with the technical requirements.Conclusion. The developed seamless and reinforced technological seam lay-flat pipelines have increased strength and performance characteristics, which allows you to improve surface irrigation methods and ensure the efficiency of water distribution across the field.

Investigation of the Degree of Augmentation of the Mass Transfer Coefficient of the Heat Transfer Medium in a Vortex Heat Exchanger of a Gas Pressure Regulating and Metering Station Heating System

Purpose of research. is to investigate the degree of augmentation of the mass transfer coefficient of a heat transfer medium in contact with a "spot" of liquid on the surface of the vortex blade when it is bombarded with dispersed contaminants in a vortex heat exchanger in order to identify a pattern that allows obtaining design values of the heat transfer coefficient of the heat transfer medium that have the best agreement with the experimental values provided in previously published articles [4, 6, 7].Methods. A complex analysis of the degree of augmentation of the mass transfer coefficient of the heat transfer medium on the surface of the vortex blade in a vortex heat exchanger based on the known theoretical positions and equations of heat and mass transfer processes.Results. The dependence of the augmentation of the mass transfer coefficient of the heat transfer medium in contact with the "spot" of liquid on the surface of the vortex blade when it is bombarded with dispersed contaminants was obtained, which allows obtaining the best agreement of the design and experimental values of the heat transfer coefficient in the vortex heat exchanger of a gas pressure regulating and metering station.Conclusion. The values of the heat transfer coefficient of the heat transfer medium calculated using the obtained dependence of the augmentation of the mass transfer coefficient of the heat transfer medium have a satisfactory convergence with the experimental data, which allows us to use this dependence in engineering calculations of the design parameters of the vortex heat exchanger used as a heat exchanger for the heating system of the working area of the gas pressure regulating and metering station. This technical solution allows not only saving natural gas as a source of heat generation, but also reducing the negative impact on the environment, since there is no need to burn natural gas. In this case, the production of thermal energy is carried out due to a regulated pressure drop of natural gas coming from the main line to consumers.

Mathematical Model for Automated Heat Flow Control of an Energy­Efficient Ventilation System

Purpose of research. In modern ventilation and air conditioning systems (VAC), one of the main components is an automatic control system (ACS) which performs various functions and also provides highly efficient operation in the range from shutdown functions to centralized regulation and control of climate parameters (temperature, humidity, monitoring concentrations of harmful substances, air speed). The goal is to study a mathematical model of heat flow control of a supply and exhaust ventilation system with a builtin integrated recuperative heat exchanger for the purpose of utilization of low-temperature heat of ventilation gases and emissions with the associated production of thermoelectricity.Methods. To achieve these goals, we used methods of mathematical simulation and computational model development. The automatic control of VAC is based on the principle of feedback – regulation of processes by obtaining information from external sensors based on mathematical simulation of physical processes occurring in the building or structure serviced.Results. An experimental supply and exhaust system with a plate heat exchanger-recuperator operates in a quasi-steady heat transfer mode. Exhaust air removed from the room is used as a heating medium. At the same time, the system is controlled using an independent scheme of connection to the heat supply system. The air heated in the room is considered as an incompressible gas, the heat exchange between the heating and heated heat transfer media is a steady­state process, the turbulence of the heating and heated flow of heat transfer media is isotropic. The result of the study is a mathematical model of heat flow control in the supply and exhaust ventilation system with a builtin integrated heat exchanger-recuperator. The optimal values of the heat energy consumed and the parameters of the ventilation system operation are obtained.Conclusion. A mathematical model of heat flow control in a supply and exhaust ventilation system with a builtin integrated heat exchanger-recuperator is proposed and investigated. The optimal values of the consumed heat energy and the parameters of the ventilation system operation are obtained.

Use of Spectral Landscape Indices for Obstacle Detection in the Tasks of Mobile Robotic Platforms Navigation in Agricultural Areas

Purpose or research is to develop an algorithm for detecting obstacles on the orthophotomap based on the analysis of the spectral landscape indices in the tasks of mobile robotic equipment navigation in agricultural areas.Methods. The following landscape indices characterizing objects of various types on a map obtained by spectral aerial photography have been considered in the paper: normalized difference vegetation index (NDVI), normalized building difference index (NDBI), normalized difference water index (NDWI), and soil-adjusted vegetation index (SAVI). These indices provide an assessment of the four main classes of objects on the map: vegetation, buildings, water bodies, and soil cover. An algorithm that provides the segmentation of zones on the map which are impassable for ground robotic means using multispectral images and the considered indices was proposed.Results. Each image is presented in the form of a colour map based on the pixel-by-pixel calculation of the indicated indices. In this case, three indices, i.e. SAVI, NDWI, NDBI, are combined (superimposed on each other), and then the NDVI layer is subtracted from the resulting image to highlight the passable zones. Thus, a formula to obtain a mask of obstacles in the image was obtained. Hence, this algorithm allows generalizing the results of calculations for all selected indices and constructing a mask of obstacles in the image. For quantitative assessment the of the algorithm execution, the area of obstacles was calculated using the indices on a sample of manually marked images. The experiments conducted show that the developed algorithm provides, on average, detection of 85.47 % of the area of all impassable zones in the images in the above classes of land cover.Conclusion. An algorithm for the automated detection of obstacles on a map obtained from a spectral orthophotomap of the area for use in the tasks of mobile robotic equipment navigation in agricultural areas has been developed and tested. In the further research, to determine flat soil areas, it is planned to modify the developed solution using the improved modified soil-adjusted vegetation index (MSAVI).

Wear Model of Sliding Rough Surfaces

Purpose of research. To develop a wear model that takes into account the quality parameters of the surface layer and friction conditions, which will allow, through controlled technological action, providing the required wear intensity of friction surfaces.Methods. Electromechanical processing, in particular, can provide the required intensity of friction surfaces wear being carried out in finishing and strengthening modes.An increase in wear resistance, endurance limit and other operational properties, and at the same time a change in the physical, mechanical and geometric parameters of the surface layer of parts is achieved by a combination of thermal and force influences on the surface of the processed part. To achieve this effect, an original installation has been developed, which includes a device for machining based on a universal machine with a set of necessary tools and devices for fixing a workpiece and supplying coolant and electric current of high power and low voltage, as well as an industrial electric current converter and control units for processing modes, switching devices and computers.Results. An approach to the representation of the nature of friction and wear has been proposed. Based on the proposed wear model, which takes into account the quality parameters of the surface layer and friction conditions, it is possible to provide the required wear intensity of friction surfaces through controlled technological action.The developed wear model, which takes into account the quality parameters of the surface layer and friction conditions, makes it possible to provide the required wear intensity of friction surfaces through controlled technological action.Conclusion. The calculated data obtained with the help of the developed model and the data obtained experimentally will find practical application in creating resource-saving processes for processing metal alloys and composite materials.