Bogdan Blagojevic

Prediction of cavitation vortex dynamics in the draft tube of a francis turbine using radial basis neural networks

Application of radial basis neural networks (RBNN) for prediction of cavitation vortex dynamics in a Francis turbine draft tube is presented. The dynamics of the cavitation vortex was established by fluctuations of a void fraction in a selected region of the draft tube. The void fraction was determined by image acquisition and analysis. Pressure in the draft tube and images of the cavitation vortex were acquired simultaneously for the experiment. RBNN were used for prediction. The void fraction in the selected region of the cavitation vortex was predicted on the basis of experimentally provided pressure data. The learning set consisted of pressure – void fraction pairs. The prediction consisted in providing only the pressure. Regression coefficients r between the predicted and measured void fractions were in an interval of 0.82–0.98. A good agreement between power spectra and correlation functions of measured and predicted void fractions was shown.

Identification of the dynamic properties of temperature-sensors in natural and petroleum gas

Abstract This paper presents a theoretical and experimental investigation of the dynamic properties of contact temperature-sensors in air as well as in natural and petroleum gas. Particularly with gas-supply systems, which can operate either with natural or petroleum gas, a knowledge of a sensor’s dynamic properties and sensitivity to different pressure and velocity conditions can be a crucial factor in the regulation and control of the system. The frequency characteristics of contact temperature-sensors in dynamic processes with air were analysed. The experimental results are presented for real Pt-resistance temperature-sensors, with an evaluation of the time-dependent uncertainty from different uncertainty sources. In real gas-supply systems, strict protective measures and a hazardous environment tend to prevent any standard experimental identification of the dynamic properties of temperature-sensors. This was the reason for our theoretical approach to the identification of the dynamic properties of the investigated temperature-sensors in natural and petroleum gas with the use of a computational calibration model, which was experimentally verified in an air-flow. Our experimental and theoretical investigation has shown that different gas compositions, pressures and velocities in the gas-supply system result in considerable variations which could lead to the applicability of the procedure in evaluating dynamic properties in highly hazardous systems.

A one-dimensional numerical model of heat and mass transfer in air-water droplet flow

This paper presents a one-dimensional numerical model of the heat, mass and momentum transfer between water droplets and humid air in the disperse process device. The model merges a model of the disintegration of a water sheet presented by Dombrowski et al. [1] and a one-dimensional numerical model of the transport phenomena in the air washer presented by Demirdžić et al. [10]. Numerical results are compared with the available experimental results. Numerical simulations can predict one-dimensional analysis of the geometrical and thermophysical parameters in the disperse process device.ZusammenfassungEin numerisches Modell für die eindimensionale Impuls-Wärme-und Stoffübertragung zwischen Wassertropfen und feuchter Luft in einer Zerstäuberapparatur wird vorgestellt. Das Modell umfaßt zwei Konzepte: (1) Zerfall einer Wasserlamelle nach Dombrowski et al [1]. (2) Eindimensionales Modell für den Transportmechanismus im Luftwäscher nach Demirdžić et al [10]. Die Validierung der numerischen Simulationen, welche die eindimensionale Analyse einzelner geometrischer und thermophysikalischer Parameter der Zerstäuberapparatur ermöglichen, erfolgte durch Vergleich mit experimentell belegten Resultaten aus der Literatur.

Simultaneous study of pressure pulsation and structural fluctuations of a cavitated vortex core in the draft tube of a Francis turbine

The paper represents a simultaneous structural dynamics analysis and static pressure pulsation of a cavitated vortex core in the draft tube of a Francis turbine. Comparison of the static pressure oscillation signal and the dynamics of the cavitated vortex core structure, which were quantified by the computer-aided visualisation led to the experimental modelling of the cavitated vortex core. Through the wavelet analysis significant correlation of the observed processes and their interaction was confirmed. The obtained results of the experiments enables us to create a numerical prediction of the cavitated vortex core dynamics based on the well-known Fanelli differential equation [1]. Measured pressure and structure fluctuations of the cavitated vortex core were introduced into the differential equation. Consequently, using the method of least squares, the characteristic coefficients of the model at different operating conditions in a Francis model turbine could be determined. Appropriate regression coefficient with its typically high values (r2 > 0.7) confirms the correct choice of the applied type of differential equation and gives the opportunity to set up a phenomenological model that shows the interdependence between the complex cavitated vortex core dynamics and the performance characteristics of a turbine.

Monitoring and Control of Quality of the Primary Layer of Mineral Wool on a Disc Spinning Machine

Abstract A novel method of measurement and control of the primary layer quality in mineral wool production processes is presented. The measurement method is based on image acquisition and processing of the mineral wool primary layer structure. As a statistical estimator of the primary layer quality, the ratio between the RMS (root mean square) and the average intensity in a selected region of the image was used. A regression model of mineral wool production process control is presented. It relates the statistical estimator of the primary layer homogeneity with the key process parameters. The regression model was derived using dimensional analysis and selection of characteristic numbers. The parameters of the regression model were selected using experimental data from a spinning machine. The results show very good agreement with a regression coefficient r 2 higher than 0.81. The regression model permits mineral wool production process control, since all the key parameters in the model can be measured in a real production process.

Thermovision method for diagnostics of local characteristics of natural draft cooling towers

Abstract Increased and more rigorous demands for power station operation in order to minimize both the cost of electrical energy production and environment pollution, addressed the necessity for optimisation of all segments of the process. As a consequence, there are certain needs for the introduction of new experimental methods which would ensure effective diagnostics. A newly developed experimental method for thermovisional detection of the temperature field in natural draft cooling towers is presented in this paper. The method is adapted to the real conditions that are present inside cooling towers and enables spatial‐ and time‐dependent detection of local temperatures in the region of drift eliminators of natural draft cooling towers. Combined with the previously developed methods for detection of velocity and temperature fields, the thermovision method enables quick detection of the local efficiency of cooling towers. Results of the method can be used for diagnostics of local and integral characteristics of cooling tower operation. Apart from this, they enable prediction of local corrections in order to increase the cooling tower efficiency.

Prediction of the Dynamic Properties of Temperature Sensors in Arbitrary Gases

Abstract This paper presents a theoretical investigation with experimental model verification of the dynamic properties identification of contact temperature sensors in arbitrary gases. For prediction of dynamic behaviour of temperature sensors in mainly uncommon atmospheres and under conditions where standard experimental work could not be carried out, a mathematical model was built for calculating the dynamic parameters of temperature sensors in various gases at different thermodynamic conditions on the basis of experimental work. Frequently, a sensors dynamic properties and sensitivity to different or changed thermodynamic conditions such as temperature, pressure, velocity, or even composition of gases in operation conditions can be a crucial factor in the regulation and control of the system. The main part of the contribution describes the mathematical model, its connection with the experimental results and limits within which the discussed relation are valid. The contribution also introduces, on the based of non-dimensional correlation and supposition that the time behaviour, is a function of pressure and velocity; a simplified expression is offered for the calculation the heat transfer coefficient as one of the most significant factor in determining the dynamic properties and behaviour of temperature sensor. The presented investigation has shown applicability of an elaborated mathematical model in connection with experimental results, as well as the employability of a simplified calculated expression which was verified with a standard iterative method. The built model and results of calculations show high applicability of the model, especially in investigations and analyses where different arbitrary gases under different thermodynamic conditions are used.

Experimental modeling of a cavitation vortex in the draft tube of a francis turbine using artificial neural networks

Experimental modeling of a cavitation vortex structure in a Francis turbine draft tube is presented. Pressure in the draft tube and images of vortex structure were acquired simultaneously for the experiment. Non-parametric radial basis neural networks were used for the experimental modeling. Two variables were modeled: average image intensities in the selected region, and entire images of the cavitation vortex. Both variables were modeled on the basis of experimentally provided pressure data. The learning set consisted of pressure–image pairs. The modeling consisted in providing only pressure, while average image intensities and images of the cavitation vortex were modeled. Regression coefficients r w between the modeled and measured average image intensities in an interval of 0.82–0.99 confirmed the correct choice of the experimental modeling method. The entire image modeling of the cavitation vortex also gave encouraging results with regression coefficients in an interval of 0.59–0.88. The regression coefficient of the entire image modeling was lower than the regression coefficient r w of the average image intensity modeling due to the high dimensionality of the cavitation vortex structure

Density and viscosity of the silicate melts for the production of the mineral wool fibres

The density and viscosity of the silicate melts are very important in the production of mineral wool because it is affected by the magnitude of fibre thickness, its magnitude depends on the kinematical and thermodynamic parameters of the real process, and on the chemical composition of raw materials used for the production of mineral wool. Density and viscosity also depend on the temperature of the melts. This article presents several known as well as unknown equations for calculation of density and viscosity of multicomponent silicate liquids. These equations are convenient for direct application in production processes.

MODELLING OF THE IMMERSION-DEPTH EFFECT IN A DRY-WELL TEMPERATURE CALIBRATOR USING AN EXPERIMENTAL DESIGN

This paper describes an investigation in the design of an improved model of an immersion-depth contact temperature sensor for dry-well heated block temperature calibrators, using experimental design methods. The obtained experimental results exhibit good agreement with the exponential regression model for measured temperature as a result of an appropriate immersion depth of the temperature sensor probe and optimal temperature of the dry-well calibrator developed by using a one-dimensional analysis of the heat transfer through the stem of the sensor probe. It is shown that, use with a dry-well temperature calibrator, which applies a much smaller number of measurements than the standard calibration procedure, reduces the total measurement uncertainty considerably. It is also important to note the correlation between the immersion depth of the heated block of the temperature calibrator and the total length of the sensor probe.