58 research outputs found
Traditional signal pattern recognition versus artificial neural networks for nuclear plant diagnostics
No full textThis paper introduces a new tool based on a traditional noise analysis technique for monitoring reactor components\u27 signal condition. It also presents the performance of artificial neural networks for pattern recognition to the same set of reactor signals and provides a comparison of these two techniques. Reactor pump signals from the Experimental Breeder Reactor (EBR-II) are utilized here. Collected signals such as pump power, pump speed, and pump pressure are obtained from already installed sensors in the reactor. The signals utilized are collected signals as well as generated signals simulating the pump shaft degradation progress. From the study of time series analysis and regression modeling of these signals, a parameter related to degradation and material buildup in the shaft is identified and used in the development of a monitoring tool. The results are then used as a benchmark against which to test the performance of artificial neural networks as a tool for reactor diagnostics. Several neural networks are examined in this study, including Restricted Coulomb Energy (RCE), Cascade Correlation, and Backpropagation paradigms of artificial neural networks. RCE is selected due to its unique design and speed, Backpropagation is selected because it is widely used and well accepted in the neural network research community, and Cascade correlation is selected because it overcomes some of the problems associated with the Backpropagation paradigm. Similar study is performed using the Adaptive Resonance Theory (ART) family of neural network paradigms. The results of this study indicate that artificial neural networks are simpler techniques for pattern recognition than noise analysis techniques such as the one introduced here. Neural networks do not require prior fault related parameter identification; they generate their own rules by learning from being shown examples. On the other hand, noise analysis and regression modeling can provide very sensitive techniques for monitoring of a detected problem in a component
Sensor Signal Analysis By Neural Networks For Surveillance In Nuclear Reactors
No full textThe application of neural networks as a tool for reactor diagnostics is examined here. Reactor pump signals utilized in a wear-out monitoring system developed for early detection of the degradation of a pump shaft [17] are analyzed as a semi-benchmark test to study the feasibility of neural networks for monitoring and surveillance in nuclear reactors. The Adaptive Resonance Theory (ART 2 and ART 2-A) paradigm of neural networks is applied in this study. The signals are collected signals as well as generated signals simulating the wear progress. The wear-out monitoring system applies noise analysis techniques and is capable of distinguishing these signals apart and providing a measure of the progress of the degradation. This paper presents the results of the analysis of these data and provides an evaluation on the performance of ART 2-A and ART 2 for reactor signal analysis. The selection of ART 2 is due to its desired design principles such as unsupervised learning, stability-plasticity, search-direct access, and the match-reset tradeoffs. ART 2-A is selected for its speed. Two simulators are built. One is ART 2, and the other ART 2-A. The result is a success for both paradigms, and the study shows that ART 2-A is not only able to learn and distinguish the patterns from each other, its learning speed is also extremely fast despite the high-dimensional input spaces. © 1992 IEE
Recommended from our members
Diagnostics and Control of Natural Gas-Fired furnaces via Flame Image Analysis using Machine Vision & Artificial Intelligence Techniques
No full textA new approach for the detection of real-time properties of flames is used in this project to develop improved diagnostics and controls for natural gas fired furnaces. The system utilizes video images along with advanced image analysis and artificial intelligence techniques to provide virtual sensors in a stand-alone expert shell environment. One of the sensors is a flame sensor encompassing a flame detector and a flame analyzer to provide combustion status. The flame detector can identify any burner that has not fired in a multi-burner furnace. Another sensor is a 3-D temperature profiler. One important aspect of combustion control is product quality. The 3-D temperature profiler of this on-line system is intended to provide a tool for a better temperature control in a furnace to improve product quality. In summary, this on-line diagnostic and control system offers great potential for improving furnace thermal efficiency, lowering NOx and carbon monoxide emissions, and improving product quality. The system is applicable in natural gas-fired furnaces in the glass industry and reheating furnaces used in steel and forging industries
Application of ART2-a as a Pseudo-supervised Paradigm to Nuclear Reactor Diagnostics
No full textAdaptive Resonance Theory (ART) represents a family of neural networks each having its own unique characteristics. This paper demonstrates the capability of ART2-A network in performing the challenging task of pattern recognition of complex noisy signals from nuclear plant components. In addition, its capability in pattern recognition of acoustic signature is briefly addressed. The results show that an ART2-A network can be successfully used both as an unsupervised pattern classifier and as a pseudo-supervised network for fault identification in a nuclear reactor system
Progress report no. 7
No full textStatement of responsibility on title-page reads: editor: M.J. Driscoll; contributors: D.C. Aldrich, M.J. Driscoll, O.K. Kadiroglu, S. Keyvan, H.U.R. Khan, D.D. Lanning, R. Morton, J. Pasztor, T.J. Reckart, A.A. Salehi, J.I. Shin, A.T. Supple, D.J. Wargo, and S.S. WuIncludes bibliographical referencesProgress report; September 30, 1976U.S. Atomic Energy Commission contracts: E(11-1) 225
Nuclear Fuel Pellet Inspection using Artificial Neural Networks
No full textNuclear fuel must be of high quality before being placed into service in a reactor. Fuel vendors currently use manual inspection for quality control of fabricated nuclear fuel pellets. In order to reduce workers\u27 exposure to radiation and increase the inspection accuracy and speed, the feasibility of automation of fuel pellet inspection using artificial neural networks (ANNs) is studied in this paper. Three kinds of neural network architectures are examined for evaluation of the ANN performance in proper classification of good versus bad pellets. Two supervised neural networks, back-propagation and fuzzy ARTMAP, and one unsupervised neural network called ART2-A are applied. The results indicate that a supervised ANN with adequate training can achieve a high success rate in classification of fuel pellets. © 1999 Elsevier Science B.V. All rights reserved
Computer-Based Teaching and Assessment in Topics on Basic Physics
No full textThis paper describes an experience in computer-based teaching and assessment in three topics in basic physics. A module is developed for each topic using the Authorware courseware authoring tool. The first module is on fundamental particles, the second on binding energy, and the third on atom density calculation. These modules are also installed on the web. Each module has four components: 1) definition, 2) example, 3) review questions, and 4) quiz. Students can see their performance on review questions interactively and have the option to repeat them, and receive on-line feedback on their score. Similarly, their performance on a quiz is evaluated on-line and feedback is provided to them. In addition, their score on each quiz as well as the time they spent taking the quiz are sent back to the instructor and stored in a permanent file. The courseware provides an overall assessment, in graphical format, of the average performance of all students who took a quiz, as well as each individual student\u27s performance. These modules are taught as supplementary part s of a course in Fundamentals of Nuclear Engineering at the University of Missouri-Rolla Nuclear Engineering Department. The experience has been positive with more than 80% of the students supporting the value of the interactive and self-pace learning of these modules
Feature Extraction for Artificial Neural Network Application to Fabricated Nuclear Fuel Pellet INSPECTION
No full text- …
