The dissertation work is devoted to the development of theoretical foundations, methodological apparatus and hardware and software tools for controlling of air pollution by energy objects based on spatially branched wireless sensor networks that will meet modern requirements for speed and level of information content, with the localization of pollution sources. Object of research: the process of air pollution control by energy facilities and the localization of pollution sources. Purpose: development of scientific and technical foundations for the construction, practical use and research of means for controlling air pollution by energy facilities and localization of pollution sources based on spatially branched wireless low-cost sensor networks. Methods of the theory of information-measuring systems, theory of electrical instrumentation, control theory, measurement theory, theory of random processes, physical and mathematical modeling, probability theory, theory of algorithms, mathematical statistics, planning of a multifactorial scientific experiment, processing and verification of its results are used. Results: the actual scientific example problem was solved, which consists in creating theoretical foundations, developing and practical application of means for controlling air pollution by energy facilities, which ensures the localization of air pollution sources on the basis of spatially branched wireless sensor networks. Novelty: improved scientific and practical foundations for the statistical study of the impact of energy balance components on emissions of pollutants into the using assessment of their correlation coefficient with small volumes of the studied samples; for the first time, an iterative method for solving the inverse problem of the distribution of pollutants in the air, namely, localizing the source of pollution and determining the concentration of its emissions, was proposed using the developed optimization models and the modified Newton method; for the first time, a method has been developed for predicting the oxygen concentration in the air, based on the analysis of meteorological parameters of the air, using the inverse functional dependence established using the discrete Fourier transform; for the first time, a matrix method for generating data from measuring modules for atmospheric air pollution control is proposed, based on detailing the properties of the generated data groups and the corresponding measures (Euclidean distance, squared Euclidean distance, Manhattan distance, Chebyshev distance or power-law distance); for the first time, a set of mathematical models is proposed for graduation of fine dust sensors, establishing a relationship between the characteristics of these measuring modules (maximum, minimum, average value, standard deviation) and the corresponding data of the reference control device (BAM device); scientific and practical aspects of assessing the technical condition of sensors of measuring modules of multichannel air pollution control networks have been developed using a number of statistical measures (determination coefficient, correlation coefficient, range and coefficient of variation) as dynamic statistical characteristics of control results; methodology for creating air pollution control systems by energy facilities, based on the modern theory of using information field models and statistical data processing methods, was further developed. The results have been implemented at SE "Plant" Elektrotyazhmash "", LLC "Ukrekoconsult"; Bud-Bud LLC, Advanced Membrane Technologies LLC; LLC "Science Park of the State Ecological Academy of Postgraduate Education and Management" CHERNOBYL""; National Aviation University and National Technical University "Kharkiv Polytechnic University".
