Ruđer Bo嗅ović Institute



Project title:

Computational Intelligence Methods in Measurement Systems

Financial support:

Ministry of Science Education and Sport


Ruđer Bo嗅ović Institute


Division of Electronics


Laboratory for Information Systems


Bijenička c. 54, Zagreb, Republic of Croatia

Contract No.:


Date of signature:

January 2, 2007


Ivan Marić, Ph.D., project leader

Dragan Gamberger, Ph.D.

Tomislav 確uc, Ph.D.

Ivan Ivek, B.Sc., Ph.D. student



Project Abstract:


Fast, reliable and accurate measurements are the precondition for any serious experimental scientific research or industrial application. The common characteristic of the contemporary measurement system is its increasing complexity. The measuring instruments and the measurement systems are becoming the components of a global information system and offer not only simple measurements but also have embedded a high level of complexity in the analysis and interpretation of measurement results. Using distributed object computing the components of the measurement system can be deployed anywhere on the network and shared by a large number of applications. The artificial intelligence methods offer the new approaches in the design of smart measurement systems.


The goal of this project is to investigate the possibilities of development and application of the artificial intelligence methods and machine learning techniques in measurement systems ranging from the development of simple models to the synthesis of sophisticated measurement procedures and smart instruments capable not only to execute simple commands but to perform rather complex tasks. A smart instrument is expected to have integrated domain-specific knowledge and learning ability that will enable it to adapt to peculiar operating conditions and to maintain high accuracy and reliability of measurements. The research will be focused on algorithms and methods for reducing the processing time and the complexity of measurement procedures and for increasing the accuracy, flexibility and the reliability of both embedded and distributed measurement systems.


By reducing the complexity of measurement algorithms and calculation procedures while preserving a high measurement accuracy we expect to shorten their calculation time thus enabling its efficient application in real-time measurements. We hope to discover accurate and reliable models of complex measurement methods and to develop adaptive instruments and the advanced measurement procedures of distributed measurement system. The proposed research is expected to solve some specific real-time flow measurement problems. We expect the cooperation with research laboratories and industry. 




Scientific papers in CC journals:


Marić, I., Optimization of self-organizing polynomial neural networks, // Expert Systems With Applications, DOI: 10.1016/j.eswa.2013.01.060, vol. 40, no 11, September, 2013, 4528-4538.


Marić, I., Ivek, I. Self-Organizing Polynomial Networks for Time-Constrained Applications. // IEEE transactions on industrial electronics, DOI: 10.1109/TIE.2010.2051934, vol. 58, no. 5, May 2011, 2019-2029.


Marić, I., Ivek, I.: Compensation for Joule傍homson effect in flowrate measurements by GMDH polynomial. // Flow measurement and instrumentation. 21 (2010), 2; 134-142



Trontl K, 確uc T, Pevec D. Support vector regression model for the estimation of g-raybuildup factors for multi-layer shields, // Annals of Nuclear Energy, 34 (2007), 12; 939-952



Marić I. A procedure for the calculation of the natural gas molar heat capacity, the isentropic exponent, and the Joule-Thomson coefficient. // Flow Measurement and Instrumentation. 18 (2007), 1; 18-26.




Scientific papers in other journals:


Trontl, K. Pevec, D. 確uc, T. Machine learning of the reactor core loading pattern critical parameters. // Science and Technology of Nuclear Installations. (2008); 695153-1-695153-7 (scientific paper).



Book chapters:


Marić, I., Ivek, I. Natural gas properties and flow computation // Natural gas, ISSN:978-953-307-112-1, SCIYO, 2010. 501-529.



Trontl, K., 確uc, T., Pevec, D.,: Learning Support Vector Regression Models for Fast Radiation Dose Rate Calculations // Machine Learning Research Progress / Peters, Hannah ; Vogel, Mia (ur.). New York : Nova Science Publishers, Inc., 2010. 427-462.




Scientific papers in conference proceedings:


Bogunović, N., 確uc, Tomislav. Applicability of Qualitative ECG Processing to Wearable Computing // Proceedings of the 5th International Workshop and Symposium on Wearable and Implanzable Body Sensor Networks / Zhang, Yuan-ting (ur.). Hong Kong : IEEE, 2008. 133-136



Trontl, K., Pevec, D., 確uc T. On Input Vector Representation for the SVR Model of Reactor Core Loading Pattern Critical Parameters // 7th International Conference on Nuclear Option in Countries with Small and Medium Electricity Grids - Conference Proceedings / Čavlina, Nikola ; Pevec, Dubravko ; Bajs, Tomislav (ur.). Zagreb : Croatian Nuclear Society, 2008. S-06.90-1-S-06.90-10



Trontl, K.; Pevec, D.; 確uc, T. Machine Learning of the Reactor Core Loading Pattern Critical Parameters // Proceedings of the International Conference Nuclear Energy for New Europe 2007. Ljubljana, Slovenia: Nuclear Society of Slovenia, 2007. 113.1-113.10.




Abstracts in book of abstracts:


Ivek, I. GMDH Structures in Time-series Modeling for Prediction // Book of Abstracts - KDSA 2008, Workshop on Knowledge Discovery in Scientific Applications / Gamberger, Dragan (ur.).
Zagreb : IRB, 2008. (presentation, abstract, scientific).


Marić, I. GMDH: building self-organizing feedforward perceptron-like polynomial models for real-time applications // Book of Abstracts - KDSA 2008, Workshop on Knowledge Discovery in Scientific Applications / Dragan Gamberger (ur.). Poreč, Hrvatska : IRB, 2008. (presentation, abstract, scientific).







Ivek, Ivan: wGMDH, Weka addon featuring GMDH, project sponsored by the Ministry of Science Education and Sports of the Republic of Croatia, 2009/2010.


I. Maric: GMDH system for desktop computer, MS Visual Studio, RBI, 2008/2009