The problem of energy load forecasting has emerged as an essential area of research for electrical distributors seeking to minimize costs, as this problem has a high degree of complexity because of modeling of human behavior. This paper solves the problem of short-term load forecasting for a day ahead using an adaptive fuzzy model, defined across the entire input space in order to share information between different areas. The proposed solution first separates the forecasting of daily load profiles into smaller, simpler subproblems, which are solved separately using a Takagi-Sugeno fuzzy model. This is done in order to solve smaller subproblems better, which results in improved forecasting accuracy after combining the subproblem results. The identification of the model is based on a recursive Gustafson-Kessel clustering and recursive weighted least mean squares, to which a combined membership function is proposed in order to improve domain partitioning. The model was tested on the real data obtained from a large Slovenian energy distribution company, at which the developed model forecast outperformed other methods, especially at the beginning of the week and during the winter. In the area of identification and implementation of fuzzy model, the research group published 4 papers in the category of the best journal from the area, the exceptional research results (A''), 11 high quality papers (A') and 3 important papers (A1/2).The research results from this area are all together presented in 111 works in SICRIS database.
COBISS.SI-ID: 11941972
This paper presents a new approach to design of experiments (DoE), based on an evolving fuzzy model structure and a supervised hierarchical clustering algorithm. DoE is the field that deals with the problem of how to design the most optimal and economic experimentation. The goal is to identify a highly nonlinear and possibly high-dimensional system, together with the minimal experimental effort required. In the area of evolving fuzzy models the research group published 2 papers in the category of the best journal from the area, the exceptional research results (A'') and 7 high quality papers (A'). The research results from this area are all together presented in 80 works in SICRIS database.
COBISS.SI-ID: 10639700
In this paper, the idea of using an evolving method as a base for the faultdetection/monitoring system is tested. The system is based on the evolving fuzzy model method. This method allows us to model the nonlinear relations between the variables with the Takagi-Sugeno fuzzy model. The method uses basic evolving mechanisms to add and remove clusters and the adaptation mechanism to adapt the clusters' and local models' parameters. In the area of process monitoring based on fuzzy models, the research group published 1 paper in the category of the best journal from the area, the exceptional research results (A'') and 2 high quality papers (A'). The research results from this area are all together presented in 16 works in SICRIS database.
COBISS.SI-ID: 10889556
The paper deals with the control of differentially driven wheeled mobile robots. Two families of wheeled mobile robots are considered: those that are capable of forward motion only and those that can perform forward and backward motion. A unified framework for the control law analysis and design of both robot types is proposed. The control laws are developed within a Lyapunov stability analysis framework. Periodic Lyapunov functions are proposed, and the constructive procedure leads to periodic control laws. In the area of mobile robotics systems the research group published 1 paper in the category of the best journal from the area, the exceptional research results (A''), 1 high quality paper (A') and 4 important papers (A1/2). The research results from this area are all together presented in 44 works in SICRIS database.
COBISS.SI-ID: 10227284
Building-management systems (BMSs) are becoming increasingly important as they are an efficient means to having buildings that consume less energy as well as for improving the indoor working and living environments. On the other hand, implementing automated control and monitoring systems in buildings is still relatively new, and one of the obstacles for their wider implementation is the ease of setting up the appropriate parameters for the controllers. During our work on an experimental controller for an indoor environment that is installed in an occupied office in the building of the Faculty of Civil and Geodetic Engineering, University of Ljubljana, Slovenia, it has become evident that a computer simulator of the system would be a welcome aid for the optimization of its functioning. In this paper we present a simulator application developed in a combined Matlab/Simulink and Dymola/Modelica environment. The simulator mirrors the functioning of the control system and the dynamics of the indoor environment, where the thermal model of the simulator was developed in the Dymola/Modelica environment, while the illuminance model was developed and parameterized as a black-box model on the basis of measurements in the Matlab environment. The simulator can emulate the response of conventional ON/OFF controllers as well as fuzzy controllers. The paper presents the design of the simulator with all of the key elements described. The underlying models for the thermal and illuminance control are also separately described. Finally, the performance of the simulator is presented for a selected day. In the area of simulation for building automation the research group published 1 paper in the category of the best journal from the area, the exceptional research results (A'') and 2 high quality paper s(A'). The research results from this area are all together presented in 19 works in SICRIS database.
COBISS.SI-ID: 10062676
This paper deals with the problem of fuzzy nonlinear model identification in the framework of a local model network (LMN). A new iterative identification approach is proposed, where supervised and unsupervised learning are combined to optimize the structure of the LMN. For the purpose of fitting the cluster-centers to the process nonlinearity, the Gustafsson–Kessel (GK) fuzzy clustering, i.e., unsupervised learning, is applied. In combination with the LMN learning procedure, a new incremental method to define the number and the initial locations of the cluster centers for the GK clustering algorithm is proposed. Each data cluster corresponds to a local region of the process and is modeled with a local linear model. Since the validity functions are calculated from the fuzzy covariance matrices of the clusters, they are highly adaptable and thus the process can be described with a very sparse amount of local models, i.e., with a parsimonious LMN model. The proposed method for constructing the LMN is finally tested on a drug-absorption-spectra process and compared to two other methods, namely, Lolimot and Hilomot. The comparison between the experimental results when using each method shows the usefulness of the proposed identification algorithm. In the area of clustering the research group published 2 papers in the category of the best journal from the area, the exceptional research results (A''), 3 high quality papers (A') and 1 important paper (A1/2). The research results from this area are all together presented in 21 works in SICRIS database.
COBISS.SI-ID: 8730708
In this book some modeling frameworks are introduced first, which can encompass the most frequently encountered complex dynamical phenomena and are practically applicable in the proposed predictive control approaches. Furthermore, unsupervised learning methods that can be used for complex-system identification are treated. Finally, several useful predictive control algorithms for complex systems are proposed and their particular advantages and drawbacks are discussed. The presented modeling, identification and control approaches are complemented by illustrative examples. The book is aimed towards researches and postgraduate students interested in modeling, identification and control, as well as towards control engineers needing practically usable advanced control methods for complex systems. In the area of control and predictive control the research group published 3 papers in the category of the best journal from the area, the exceptional research results (A''), 14 high quality papers (A') and 13 important papers (A1/2). The research results from this area are all together presented in 45 works in SICRIS database.
COBISS.SI-ID: 9548628
Original research paper in the SCI journal, describing the upgrades and enhancements of the existent mathematical models of the electric arc furnace (EAF) processes, i.e. thermal, chemical and mass-transfer. The model is developed according to fundamental physical laws and is validated on operational measurements of the EAF. The model is intended to be used in simulation studies, for control design purposes and for optimization of the steel-melting process. In the area of modelling, identification and optimization of EAF the research group published 10 important papers (A1/2). The research results from this area are all together presented in 13 works in SICRIS database. The research from this area is patented in: GB 2507116 B, 2014-04-23 Soft sensor for online estimation of the steel bath temperature in an electric furnace (EAF). Research on modelling of EAF process has been recognized as an outstanding scientific achievement and the work has been presented at the Excellent in Science 2013, an event that is organized by the Slovenian research Agency (ARRS). The model is protected by GB patent.
COBISS.SI-ID: 11444820
The book covers the main topics from the wide area of wheeled mobile robotics, explaining all applied theory and applications. The book gives the reader a good foundation, enabling them to continue to more advanced topics. Several examples are included for better understanding, many of them accompanied by short MATLAB script code making it easy to reuse in practical work. The book includes several examples of discussed methods and projects for wheeled mobile robots and some advanced methods for their control and localization. It is an ideal resource for those seeking an understanding of robotics, mechanics, and control, and for engineers and researchers in industrial and other specialized research institutions in the field of wheeled mobile robotics. Beginners with basic math knowledge will benefit from the examples, and engineers with an understanding of basic system theory and control will find it easy to follow the more demanding fundamental parts and advanced methods explained. In the area of mobile robotics systems the research group published 1 paper in the category of the best journal from the area, the exceptional research results (A''), 1 high quality paper (A') and 4 important papers (A1/2). The research results from this area are all together presented in 44 works in SICRIS database.
COBISS.SI-ID: 11671636
This paper presents the design and implementation of a unique control system for a smart hoist, a therapeutic device that is used in rehabilitation of walking. The control system features a unique human–machine interface that allows the human to intuitively control the system just by moving or rotating its body. The paper contains an overview of the complete system, including the design and implementation of custom sensors, dc servo motor controllers, communication interfaces and embedded-system based central control system. The prototype of the complete system was tested by conducting a 6-runs experiment on 11 subjects and results are showing that the proposed control system interface is indeed intuitive and simple to adopt by the user. Embedded control system for smart walking assistance device has been recognized as an outstanding scientific achievement and the work has been presented at the Excellent in Science 2017, an event that is organized by the Slovenian research Agency (ARRS). For the presented unique control system a European patent application is also pending.
COBISS.SI-ID: 11361364