Magnetic levitation systems have become very important in many applications. Due to their instability and high nonlinearity, such systems pose a challenge to many researchers attempting to design high-performance and robust tracking control. This paper proposes an improved adaptive fuzzy backstepping control for systems with uncertain input nonlinear function (uncertain parameters and structure), and applies it to a magnetic levitation system, which is a typical representative of such systems. An adaptive fuzzy system is used to approximate unknown, partially known or uncertain input nonlinear functions of a magnetic levitation system. An adaptation law is obtained based on Ljapunov analysis in order to guarantee closed-loop stability and good tracking performance. Initial adaptive and control parameters have been initialized with Symbiotic Organism Search optimization algorithm, due to strong non-linearity and instability of the magnetic levitation system. The theoretical background of the proposed control method is verified with a simulation study and implementation on a laboratory experimental application.
COBISS.SI-ID: 20349462
A different modelling approach for modelling of hydraulic systems is presented in this paper. The modelling approach deals with how to obtain adequate mathematical models of the most common hydraulic elements, which are often used in process industries. Partial models should be suitable for behaviour simulation of the complete hydraulic system, for simulation of hydraulic process control applications, for supervision of hydraulic processes, and for educational purposes. All partial models, model of the pump, model of the pipeline, model of the tank, and model of the control valve, are connected logically and properly connected in a MATLAB/Simulink environment and form the mathematical model of the desired hydraulic system. The difference between the classical and presented approach is explained where non-linear and linearized models of all hydraulic elements tanks are derived. Two hydraulic processes are simulated and results show the potential of the presented modelling approach. The proposed hydraulic modelling approach was used in Bachelor and Master Study Programmes in the scope of students’ individual work as seminar or project work. The model and real process measurements gave students a proper insight of modelling problems, design and selection issues of equipment and examples of good engineering practice. Positive students’ feedback was towards understanding the underlying physics of the developed theoretic models in connection to real process behaviour and measurements, which is mostly missing when dealing with classic modelling and simulation courses.
COBISS.SI-ID: 21136406
This paper presents a miniature, all-optical, thermal conductivity fiber-optic sensor, which can be applied to various fluid composition identification situations. The sensor is composed of a short section of highly absorbing fiber, which is configured as a Fabry-Perot interferometer. A higher power laser diode is used to heat the absorbing fiber periodically, while proper signal integration system issued in order to observe temperature variations of the heated fiber. These variations are further correlated to the surrounding fluid’s thermal conductivity. The sensor was applied experimentally to various fluid identification situations, including gas/liquid and liquid–liquid phase detection, identification of different fluids (including liquids and gases), liquids’ binary mixtures mass ratio determination, and some common fluid identification applications.
COBISS.SI-ID: 20017174
This paper deals with the implementation of a control algorithm based on a state controller with an adaptation of the state controller gains by using the fuzzy-logic approach. Adjustable state controller gains cause that the static error can be reduced arbitrarily for a system with variable parameters as is the DC–DC step-down converter for power supply systems. Fuzzy sets are tuned offline by a genetic algorithm using fuzzy-logic and a genetic toolbox in MATLAB. Fuzzification and defuzzification algorithms are implemented in real time within a Field Programmable Gate Array circuit. The whole control algorithm is performed within a sampling time of 5.33 μs. Operation of the controller is verified experimentally.
COBISS.SI-ID: 20583958
Temporal trends in source normalized impact per paper (SNIP) values for the three top-ranking nursing journals were analyzed and compared to explore whether predicting future SNIP values based on trend analysis could be an innovative service provided by librarians. The International Journal of Nursing Studies, Journal of Nursing Scholarship, and Journal of Advanced Nursing were the three top-ranked nursing journals according to 2015 SNIP values. SNIP values for the selected journals were retrieved from the Scopus database, and extracted data were exported to Joinpoint trend analysis software to perform trend analysis.Predictions of journal metrics based on statistical joinpoint regression may not be completely accurate. Using this technique, however, a librarian can reasonably claim which journal will retain or even improve its prestige in the future and thus safely advise prospective authors on where to publish their research.
COBISS.SI-ID: 20847382