Ognjen Kuljaca

Internet-based educational control systems lab using NetMeeting

The purpose of the paper is to illustrate the remote control of hardware in the laboratory for educational purposes, using commercial off-the-shelf equipment and available freeware. The goal is to eliminate extensive programming using high-level languages, such as Java programming to achieve remote connectivity and focus on the primary task of implementing control algorithms. The use of MS NetMeeting for Internet-based control is described. Control experiments were easily carried out by a remote user, using Matlab-based real-time tools and MS NetMeeting.

Speed and active power control of hydro turbine unit

In the paper, the procedure is given for designing speed and active power controller of hydro turbine units. The procedure is based on mathematical models of the controlled system. The controller parameters are obtained from closed-loop poles and hydro turbine parameters by derived analytical formulas over a wide range of the hydro turbine operating points. The described procedure allows the fast and direct determination of the controller parameters. There is no need for heuristic controller parameters tuning. Results from two hydro turbine units with the controllers designed using the described procedure are given as an illustration.

Analytical Determination of Describing Function of Nonlinear Element with Fuzzy Logic

The paper presents the procedure for analytical determination of describing function of nonlinear element with fuzzy logic nonlinearity. The procedure is based on the method of analytical determination of describing function of generalized static characteristic of nonlinear element. The procedure is also illustrated with examples.

Neural network frequency control for thermal power systems

A neural network control scheme for thermal power system is described. No off-line training is required for the proposed neural network controller. The online tuning algorithm is provided. The stability of the controller is proven. The performance of the controller is illustrated via simulation.

Engineering procedure for analysis of nonlinear structure consisting of fuzzy element and typical nonlinear element

Engineering procedure for approximate analytical determination of describing function of nonlinear systems with odd static characteristics is presented in the paper. Generalized mathematical expressions for determining such describing function with error estimation are given. The procedure is illustrated on determination of describing function, and corresponding error estimation, of nonlinear structure consisting of fuzzy element and typical nonlinear element.

Computer Aided Harmonic Linearization of SISO Systems Using Linearly Approximated Static Characteristic

Harmonic linearization, also known as describing function analysis is well known method for analysis of SISO systems. Computer aided harmonic linearization, as experimental practical method for determining static characteristics and describing functions is described. The use of linearly approximated static characteristic in determination of describing function introduces an inherent error resulting from the difference between original static characteristic and corresponding linearly approximated static characteristic. Error estimation and comparison of approximated characteristic error with the actual error is illustrated on several examples.

Exploring Bayesian networks for automated breast cancer detection

This paper gives an introduction to the Bayesian networks for the exploration of implementing a Bayesian belief network for an automated breast cancer detection support tool. It is intuitive that Bayesian networks can be employed as one viable option for computer-aided detection by representing the relationships between diagnoses, physical findings, laboratory test results, and imaging study findings. This paper brings important entities such as Radiologists, Image Processing Scientists, Data Base Specialists and Applied Mathematicians on a common platform. A brief background concerning causal networks, probability theory and Bayesian networks is given. Available computational tools and platforms are described. Steps towards building a Bayesian Belief Network Implementation are introduced.

Multisensory data exploitation using advanced image fusion and adaptive colorization

Multisensory data usually present complimentary information such as visual-band imagery and infrared imagery. There is strong evidence that the fused multisensor imagery increases the reliability of interpretation, and the colorized multisensor imagery improves observer performance and reaction times. In this paper, we propose an optimized joint approach of image fusion and colorization in order to synthesize and enhance multisensor imagery such that the resulting imagery can be automatically analyzed by computers (for target recognition) and easily interpreted by human users (for visual analysis). The proposed joint approach provides two sets of synthesized images, a fused image in grayscale and a colorized image in color using a fusion procedure and a colorization procedure, respectively. The proposed image fusion procedure is based on the advanced discrete wavelet (aDWT) transform. The fused image quality (IQ) can be further optimized with respect to an IQ metric by implementing an iterative aDWT procedure. On the other hand, the daylight coloring technique renders the multisensor imagery with natural colors, which human users are use to observing in everyday life. We hereby propose to locally colorize the multisensor imagery segment by mapping the color statistics of the multisensor imagery to that of the daylight images, with which the colorized images resemble daylight pictures. This local coloring procedure also involves histogram analysis, image segmentation, and pattern recognition. The joint fusion and colorization approach can be performed automatically and adaptively regardless of the image contents. Experimental results with multisensor imagery showed that the fused image is informative and clear, and the colored image appears realistic and natural. We anticipate that this optimized joint approach for multisensor imagery will help improve target recognition and visual analysis.

Marine Diesel Engine Process Modelling and Control Using Advanced Technologies

The paper presents some possibilities of practical use of advanced computing technologies applied to the modelling and control of marine diesel engine. The emphasis is put on two well recognised techniques, fuzzy logic and artificial neural networks. Because of the complexity of diesel propulsion engine working in changeable operating regimes and environmental conditions at sea, it is highly desirable to have adaptive control system capable of accommodating to such situations. We propose here an effective method of engine control system design using multiple engine models and adequate algorithms with learning and adapting possibilities to different operating and environmental conditions. Simulation examples applied to marine diesel engine of MAN B&W 6L60MC type show advantages of using combined methods and techniques in engine modelling and control.

Neuro-fuzzy modelling of marine diesel engine cylinder dynamics

Abstract In this paper, the practical application of some well recognised fuzzy methods and neural networks techniques to modelling marine diesel engine cylinder dynamics using real-time data and expert knowledge has been considered. The simulation was done in Matlab environment with real-time data originated from 2-stroke marine diesel propulsion engine on test bed during final testing, combined with knowledge elicited from engine experts and experienced test bed operators. Takagi-Sugeno fuzzy model has been designed based on cylinder pressure data after their clustering using fuzzy subtractive method. Model parameter tuning was investigated using ANFIS with combined learning algorithms: least-squares and back-propagation gradient descent method. The model obtained can be of practical importance in engine working regime adjustment, predicting cylinder data in faulty sensor case or adaptive threshold tuning within faults detection and identification.