Martyniuk G. Computerized system for evaluating the characteristics of noise signals for the diagnosis of wind power units

The object of research: noise signals of the state and mode of operation of wind power units. The purpose of research: the substantiation of the mathematical model of noise signals, statistical methods for processing data of computer simulation and measurements of noise signals, and the development of a computerized system for evaluating the characteristics of noise signals for the diagnosis of wind power units. Methods of research: methods of the theory of random processes and mathematical statistics, the theory of signals and systems, the theory of measurements, digital processing of signals. Theoretical and practical results: the general mathematical model of the noise signal of energy objects in the form of a three-dimensional vector random process was further developed, which made it possible to increase the efficiency of the application of information-measuring systems for the evaluation of the noise signal characteristics; for the first time on the basis of the general model of the noise signal of the energy object a constructive model of the noise signal of the energy object was proposed in the form of a three-component vector random process that can be used for computer simulation in the studies of the efficiency of the computerized system; for the first time in the study of computer simulation of noise signals, the characteristics of the quality of pseudorandom number generation methods were obtained; the method of detecting the instantaneous moments of the disturbance of the time series of noise signals is developed for obtaining the intervals of homogeneity of time series, which provides evaluation of the characteristics of noise signals at each interval of homogeneity separately; the number of components of the vector of diagnostic signs of WPU has been increased, which made it possible, together with known diagnostic features, to use in this vector the trend component of the time series, time moments of the dynamics of the change-point detection, intervals of homogeneity of the row between the moments of the change-point. This allowed to increase the information resource of the diagnostic process of the state and modes of operation of the WPU; the methodological foundations of the integrated approach to the diagnostics of energy objects have been improved on the basis of mathematical and computer modeling, statistical evaluation of the noise signal characteristics; recommendations for choosing a method for modeling pseudo-random numbers depending on the sample size of the realizations of the noise signal are obtained. They can be used to study the effectiveness of a computerized system for estimating the characteristics of noise signals; classification of statistical methods of chenge-point detection of time series dynamics of measurement data of noise signals is carried out. This provided a practical opportunity to interpret changes in the state of the modes of operation of energy objects based on the evaluation of the characteristics of noise signals; a computerized system for estimating the characteristics of noise signals for the diagnostics of wind power units was developed and its work was investigated on the basis of an experimental sample of a computerized system; the results of the conducted experimental studies confirm the effectiveness of the methods and approaches to the development of noise signals of wind power units. A subject degree of implantation: are used in the State Enterprise of the Ministry of Defense of Ukraine "Lutsk Repair Plant "Motor", and at the department of information-measuring systems of the National Aviation University. Effectiveness of implantation: significant technical and economic results are received which is to increase the efficiency indicators of the methods and tools for processing the noise signals of energy objects. Sphere of use: systems for assessing the characteristics of noise signals for a wide range of subject-matter, as well as for the development and testing of diagnostic methods for technical systems and energy objects.