V.N. Kostyukov’s research while affiliated with Omsk State Technical University and other places

A mathematical model of the working processes of a hybrid displacement piston power machine with profiled groove seal is proposed. Calculation of the working processes in the compressor section in the presence of a liquid film under the piston and a “pump” course in the pump section and a piston seal is presented. The working processes for a hybrid displacement piston power machine with a smooth groove seal and with a profiled groove seal is calculated and the influence of a profiled groove seal on the working processes and the dynamics of the layer of liquid over the piston is demonstrated.

Improvement of distinguishing criteria, determining defects of machinery and mechanisms, by vibroacoustic signals is a recent problem for technical diagnostics. The work objective is assessment of instantaneous values by methods of statistical decision making theory and risk of regulatory values of characteristic function modulus. The modulus of the characteristic function is determined given a fixed parameter of the characteristic function. It is possible to determine the limits of the modulus, which correspond to different machine's condition. The data of the modulus values are used as diagnostic features in the vibration diagnostics and monitoring systems. Using such static decision-making methods as: minimum number of wrong decisions, maximum likelihood, minimax, Neumann-Pearson characteristic function modulus limits are determined, separating conditions of a diagnosed object.

Examples of using the method developed for the earlier proposed concept of the monitoring system of the technical condition of a turbounit are presented. The solution methods of the inverse problem—the calculation of misalignments of supports based on the measurement results of positions of rotor pins in the borings of bearings during the operation of a turbounit—are demonstrated. The results of determination of static responses of supports at operation misalignments are presented. The examples of simulation and calculation of misalignments of supports are made for the three-bearing “high-pressure rotor–middle-pressure rotor” (HPR–MPR) system of a turbounit with 250 MW capacity and for 14-supporting shafting of a turbounit with 1000 MW capacity. The calculation results of coefficients of the stiffness matrix of shaftings and testing of methods for solving the inverse problem by modeling are presented. The high accuracy of the solution of the inverse problem at the inversion of the stiffness matrix of shafting used for determining the correcting centerings of rotors of multisupporting shafting is revealed. The stiffness matrix can be recommended to analyze the influence of displacements of one or several supports on changing the support responses of shafting of the turbounit during adjustment after assembling or repair. It is proposed to use the considered methods of evaluation of misalignments in the monitoring systems of changing the mutual position of supports and centerings of rotors by half-couplings of turbounits, especially for seismically dangerous regions and regions with increased sagging of foundations due to watering of soils.

The paper dwells upon urgent issues of evaluating impact of actions conducted by complex technological systems operators on their safe operation considering application of condition monitoring systems for elements and sub-systems of petrochemical production facilities. The main task for the research is to distinguish factors and criteria of monitoring system properties description, which would allow to evaluate impact of errors made by personnel on operation of real-time condition monitoring and diagnostic systems for machinery of petrochemical facilities, and find and objective criteria for monitoring system class, considering a human factor. On the basis of real-time condition monitoring concepts of sudden failure skipping risk, static and dynamic error, monitoring systems, one may solve a task of evaluation of impact that personnel’s qualification has on monitoring system operation in terms of error in personnel or operators’ actions while receiving information from monitoring systems and operating a technological system. Operator is considered as a part of the technological system. Although, personnel’s behavior is usually a combination of the following parameters: input signal - information perceiving, reaction – decision making, response – decision implementing. Based on several researches on behavior of nuclear powers station operators in USA, Italy and other countries, as well as on researches conducted by Russian scientists, required data on operator’s reliability were selected for analysis of operator’s behavior at technological facilities diagnostics and monitoring systems. The calculations revealed that for the monitoring system selected as an example, the failure skipping risk for the set values of static (less than 0.01) and dynamic (less than 0.001) errors considering all related factors of data on reliability of information perception, decision-making, and reaction fulfilled is 0.037, in case when all the facilities and error probability are under control – not more than 0.027. In case when only pump and compressor units are under control, the failure skipping risk is not more than 0.022, when the probability of error in operator’s action is not more than 0.011. The work output shows that on the basis of the researches results an assessment of operators’ reliability can be made in terms of almost any kind of production, but considering only technological capabilities, since operators’ psychological and general training considerable vary in different production industries. Using latest technologies of engineering psychology and design of data support systems, situation assessment systems, decision-making and responding system, as well as achievement in condition monitoring in various production industries one can evaluate hazardous condition skipping risk probability considering static, dynamic errors and human factor.

The paper presents the results obtained in the studies of the orthogonality of the vibration diagnostic features system comprising the integrated features, particularly – root mean square values of vibration acceleration, vibration velocity, vibration displacement and fractal feature (Hurst exponent). To diagnose the condition of the equipment by the vibration signal, the orthogonality of the vibration diagnostic features is important. The fact of orthogonality shows that the system of features is not superfluous and allows the maximum coverage of the state space of the object being diagnosed. This, in turn, increases reliability of the machinery condition monitoring results. The studies were carried out on the models of vibration signals using the programming language R.

The paper considers the relationship between the parameters of the vibration arising in rotating machinery during operation and the shaft speed. The goal of this paper is to determine the dependence of the vibration parameters on the shaft speed for solving applied engineering problems. To properly evaluate the technical condition of bearing assemblies, we should take into account the pattern of the rotating machinery vibration parameters-shaft speed relationship, which will allow creating new diagnostic features, the totality of which will ensure an increased reliability of diagnosis. We took the check for a correlation between the factor and resultative feature parameters as the correlation analysis method. A high pair linear correlation between the diagnostic features (acceleration, velocity, displacement) and the shaft speed was determined on the basis of the check for correlation between the vibration parameters and the shaft speed, and also the linear correlation coefficients can be used to solve the applied engineering problems of diagnosing the bearing assemblies of the rotating machinery.

The paper reviews the publications on low-frequency vibration in power plants. It is noted that the regulatory literature poorly defines the issues of rating and diagnosing self-excited and subharmonic vibrations. Analysis of the literature and ISO standards shows that despite the considerable experience gained by specialists in low-frequency vibration control, a number of issues, such as subharmonic resonances and nonlinear properties of complicated multi-seated rotor systems supported by journal bearings, are still understudied and require further calculations and experiments. The paper presents some results obtained in experiments of low-frequency vibrations for a rotor supported by journal bearings and having a residual deflection. The experimental results confirmed the classical hysteresis of self-excited vibrations – the difference in the boundaries of self-excited vibrations during ascent and descent is about 4 Hz. The arears of appearance and disappearance of subharmonic vibrations are shown using the spectral characteristics of vibrations and cascade spectra.

The rationale is given for the improvement of the regulatory framework for the use of shaft sensors for the in-service condition monitoring of turbo generators and the development of control systems of shaft surfacing and misalignments of supports. A modern concept and a set of methods are proposed for the condition monitoring of the “shaft line–thrust bearing oil film–turbo generator supports” system elements based on the domestic COMPACS® technology. The system raw data are design, technology, installation, and operating parameters of the turbo generator as well as measured parameters of the absolute vibration of supports and mechanical quantities, relative displacements and relative vibration of the rotor teeth in accordance with GOST R 55263-2012. The precalculated shaft line assembly line in the cold state, the nominal parameters of rotor teeth positions on the dynamic equilibrium curve, the static and dynamic characteristics of the oil film of thrust bearings, and the shaft line stiffness matrix of unit support displacements have been introduced into the system. Using the COMPACS-T system, it is planned to measure positions and oscillations of rotor teeth, to count corresponding static and dynamic characteristics of the oil film, and the static and dynamic loads in the supports in real time. Using the obtained data, the system must determine the misalignments of supports and corrective alignments of rotors of coupling halves, voltages in rotor teeth, welds, and bolts of the coupling halves, and provide automatic conclusion if condition monitoring parameters correspond to standard values. A part of the methodological support for the proposed system is presented, including methods for determining static reactions of supports under load, the method for determining shaft line stiffness matrices, and the method for solving the inverse problem, i.e., the determination of the misalignments of the supports by measurements of rotor teeth relative positions in bearing housings. The procedure for calculating misalignments of turbo generator shaft line supports is set out.