Stefan Figedy - Academia.edu (original) (raw)
Uploads
Papers by Stefan Figedy
In this work a new type of in-core sensor validation method is outlined, which is based on combin... more In this work a new type of in-core sensor validation method is outlined, which is based on combination of correlation coefficients and mutual information indices. First experience with this approach is described and further improvements to enhance the outcome reliability are proposed namely through computational intelligence techniques.
The aim of this overview paper is to present the new approaches elaborated at VUJE, Inc. to make ... more The aim of this overview paper is to present the new approaches elaborated at VUJE, Inc. to make the diagnostic systems in the environment of a nuclear power plant more robust and reliable. The systems in question are the sensor signal validation, denoising, early detection and identification of anomalies in chemical regime, time prediction of plant process parameters, and loose part monitoring in the reactor primary circuit. The techniques used are based on the computational intelligence and modern signal processing methods. The development of these systems was very much facilitated by variety of methods that can be used in MATLAB environment together with the toolboxes like Signal processing, Fuzzy Logic, Neural Networks, and Wavelet toolbox. This paper is of an overview character and does not focus on a detailed description of the methods used. 1 Real-Time Process Signal Validation Based on Neuro-Fuzzy and Possibilistic Approach The operation of each industrial plant is based on ...
This article provides brief information about the fundamental features of a newlydeveloped diagno... more This article provides brief information about the fundamental features of a newlydeveloped diagnostic system for early detection and identification of anomalies being generated in water chemistry regime of the primary and secondary circuit of the VVER-440 reactor. This system, which is called SACHER (System of Analysis of CHEmical Regime), was installed within the major modernization project at the NPP-V2 Bohunice in the Slovak Republic. The SACHER system has been fully developed in MATLAB environment. It is based on computational intelligence techniques and inserts various elements of intelligent data processing modules for clustering, diagnosing, future prediction, signal validation, etc, into the overall chemical information system. The application of SACHER would essentially assist chemists to identify the current situation regarding anomalies being generated in the primary and secondary circuit water chemistry. This system is to be used for diagnostics and data handling, howeve...
World Scientific Proceedings Series on Computer Engineering and Information Science, 2012
... A NEURO-FUZZY MODEL 307 operating region, the basic idea is to split this region in Q fuzzy c... more ... A NEURO-FUZZY MODEL 307 operating region, the basic idea is to split this region in Q fuzzy clus-ters and derive a mapping function that ... In pattern recognition the Q clusters are identified by prototype patterns, which in the case of spherical or ellipsoidal clusters are also ...
The State of the Art in Computational Intelligence, 2000
Progress in Nuclear Energy, 2005
In this work a new type of in-core sensor validation method is outlined, which is based on combin... more In this work a new type of in-core sensor validation method is outlined, which is based on combination of correlation coefficients and mutual information indices. First experience with this approach is described and further improvements to enhance the outcome reliability are proposed namely through computational intelligence techniques.
The State of the Art in Computational Intelligence, 2000
In this work a new type of in-core sensor validation method is outlined, which is based on combin... more In this work a new type of in-core sensor validation method is outlined, which is based on combination of correlation coefficients and mutual information indices. First experience with this approach is described and further improvements to enhance the outcome reliability are proposed namely through computational intelligence techniques.
The aim of this overview paper is to present the new approaches elaborated at VUJE, Inc. to make ... more The aim of this overview paper is to present the new approaches elaborated at VUJE, Inc. to make the diagnostic systems in the environment of a nuclear power plant more robust and reliable. The systems in question are the sensor signal validation, denoising, early detection and identification of anomalies in chemical regime, time prediction of plant process parameters, and loose part monitoring in the reactor primary circuit. The techniques used are based on the computational intelligence and modern signal processing methods. The development of these systems was very much facilitated by variety of methods that can be used in MATLAB environment together with the toolboxes like Signal processing, Fuzzy Logic, Neural Networks, and Wavelet toolbox. This paper is of an overview character and does not focus on a detailed description of the methods used. 1 Real-Time Process Signal Validation Based on Neuro-Fuzzy and Possibilistic Approach The operation of each industrial plant is based on ...
This article provides brief information about the fundamental features of a newlydeveloped diagno... more This article provides brief information about the fundamental features of a newlydeveloped diagnostic system for early detection and identification of anomalies being generated in water chemistry regime of the primary and secondary circuit of the VVER-440 reactor. This system, which is called SACHER (System of Analysis of CHEmical Regime), was installed within the major modernization project at the NPP-V2 Bohunice in the Slovak Republic. The SACHER system has been fully developed in MATLAB environment. It is based on computational intelligence techniques and inserts various elements of intelligent data processing modules for clustering, diagnosing, future prediction, signal validation, etc, into the overall chemical information system. The application of SACHER would essentially assist chemists to identify the current situation regarding anomalies being generated in the primary and secondary circuit water chemistry. This system is to be used for diagnostics and data handling, howeve...
World Scientific Proceedings Series on Computer Engineering and Information Science, 2012
... A NEURO-FUZZY MODEL 307 operating region, the basic idea is to split this region in Q fuzzy c... more ... A NEURO-FUZZY MODEL 307 operating region, the basic idea is to split this region in Q fuzzy clus-ters and derive a mapping function that ... In pattern recognition the Q clusters are identified by prototype patterns, which in the case of spherical or ellipsoidal clusters are also ...
The State of the Art in Computational Intelligence, 2000
Progress in Nuclear Energy, 2005
In this work a new type of in-core sensor validation method is outlined, which is based on combin... more In this work a new type of in-core sensor validation method is outlined, which is based on combination of correlation coefficients and mutual information indices. First experience with this approach is described and further improvements to enhance the outcome reliability are proposed namely through computational intelligence techniques.
The State of the Art in Computational Intelligence, 2000