ИзмеренИя выбросов золы установкИ по сжИганИю радИоактИвно загрязненной древесИны Исследована эффективность пылегазоочистного оборудования, установленного в г. Чернобыль (Украина) на инсинераторы радиоактивно загрязненной древесины КВм(а)-2,0. Данная установка включает в себя грубую и тонкую очистку (группа циклонов типа ЦП-15 и группа рукавных фильтров). В результате проведенных экспериментальных исследований подтверждена эффективность работы пылегазоочистного оборудования и достигнуты требования по выбросам инсинераторной установки за концентрацией золы до 4 мг/м 3. ключевые слова: выбросы загрязняющих веществ, экологические измерения, инсинератор радиоактивно загрязненной древесины, пылегазоочистное оборудование.

In this paper, it is shown that based on an analysis of the implemented functions of existing robot manipulators the task of automation of the safe capture of objects by a robot during the assembly process is poorly developed. In the process of analysis,there were discovered technological solutions to three main tasksfor the development of a subsystem for capturing objects by a robotic manipulator: determination of the dimensions and shape of the capturing object; determination of the distance from the robot manipulator to all the points of the capturing object, determination of the capture point of the object and clarification of the distance to the robot manipulator. It is shown that all of the above tasks are not sufficiently solved. Therefore, it was proposed to develop a methodology for creating an intelligent software and hardware subsystem for capturing an object by robot manipulator. The developed methodology consists ofsix steps: obtaininga stereo image and buildingan in-depth map; determination of the distance from the robot manipulator to all the points of the object; determination of the contour of the object; determination of the capture point of the objectand clarification of the distance to the robot manipulator; determination of the degree of capturing the object; determination of the movement of the manipulator to capture an object at the desired point. To find the capture point, it is proposed to use the contour search method on the object’s depth map, and to search for the finest part on the contour, limit it to a segment and find its middle point. To implement the algorithm for determination of the distance to the object, the degree of its capture and movement to the desired point, the dependencies of the calculations on the depth map and the physical characteristics of the manipulator are formalized. The capabilities of the StereoPi microprocessor are analyzed and its use for the hardware solution of the capture function by the robotic arm is proposed. The simulation of the intelligentsoftware and hardware subsystem for capturing an object of complex shape has been performed. Conclusions are drawn about the independence of the developed subsystem from the type of object and its viewing angle by a robot manipulator. In addition, an increase in the accuracy of capturing the object by a robot manipulator equipped with an intelligent subsystem is shown in comparison with its existing analog. Potential problems inthe implementation of the proposed methodology are highlighted.

The application of deep learning convolutional neural networks for solving the problem of automated facial expression recognition and determination of emotions of a person is analyzed. It is proposed to use the advantages of the transfer approach to deep learning convolutional neural networks training to solve the problem of insufficient data volume in sets of images with different facial expressions. Most of these datasets are labeled in accordance with a facial coding system based on the units of human facial movement. The developed technology of transfer learning of the public deep learning convolutional neural networks families DenseNet and MobileNet, with the subsequent “fine tuning” of the network parameters, allowed to reduce the training time and computational resources when solving the problem of facial expression recognition without losing the reliability of recognition of motor units. During the development of deep learning technology for convolutional neural networks, the following tasks were solved. Firstly, the choice of publicly available convolutional neural networks of the DenseNet and MobileNet families pre-trained on the ImageNet dataset was substantiated, taking into account the peculiarities of transfer learning for the task of recognizing facial expressions and determining emotions. Secondary, a model of a deep convolutional neural network and a method for its training have been developed for solving problems of recognizing facial expressions and determining human emotions, taking into account the specifics of the selected pretrained convolutional neural networks. Thirdly, the developed deep learning technology was tested, and finally, the resource intensity and reliability of recognition of motor units on the DISFA set were assessed. The proposed technology of deep learning of convolutional neural networks can be used in the development of systems for automatic recognition of facial expressions and determination of human emotions for both stationary and mobile devices. Further modification of the systems for recognizing motor units of human facial activity in order to increase the reliability of recognition is possible using of the augmentation technique.

The new method of visual diagnostics of liquid motion processes in physical models showed a high degree of the flow structure organization. Visual pictures made it possible to develop a hydraulic experiment to reveal the dimensions of the transverse structure in the form of layers and zones of flow separation from the channel walls. Visual diagnostics is the basis for comprehensive equipment design. Visual studies of the flow structure provide information for improving equipment by changing the geometry of the flow paths. Hydraulic studies show the change in the resistance of the equipment channels. Based on the results of visual and hydraulic studies, the wave character of the distribution of the pulsation velocity components was revealed. The regularities of the velocity distribution allow predicting the minimum or maximum values of the resistances of the flow paths of the equipment.

На основе анализа параметров дутьевых трактов и режимов работы вентиляторов предложен вариант снятия ограничений тепловой мощности котельных установок марки КВГМ (котел водогрейный, газомазутный) по дутью за счёт корректировки аэродинамики во входном патрубке вентилятора и вспомогательных элементах дутьевого тракта типа «поворот». Предлагаемый вариант снятия ограничений тепловой мощности является энергосберегающим потому, что позволяет существенно увеличить подачу воздуха в котельную установку с одновременным снижением затрат мощности на привод вентилятора.