Journal article Open Access

Applying the Computational Intelligence Paradigm to Nuclear Power Plant Operation: A Review (1990-2015).

Tatiana Tambouratzis T., Giannatsis J., Kyriazis A., and Siotropos P.


Citation Style Language JSON Export

{
  "DOI": "10.4018/IJEOE.2020010102", 
  "container_title": "International Journal of Energy Optimization and Engineering", 
  "language": "eng", 
  "title": "Applying the Computational Intelligence Paradigm to Nuclear Power Plant Operation: A Review (1990-2015).", 
  "issued": {
    "date-parts": [
      [
        2019, 
        12, 
        16
      ]
    ]
  }, 
  "abstract": "<p>In the guise of artificial neural networks (ANNs), genetic/evolutionary computation algorithms<br>\n(GAs/ECAs), fuzzy logic (FL) inference systems (FLIS) and their variants as well as combinations,<br>\nthe computational intelligence (CI) paradigm has been applied to nuclear energy (NE) since the<br>\nlate 1980s as a set of efficient and accurate, non-parametric, robust-to-noise as well as to-missinginformation,<br>\nnon-invasive on-line tools for monitoring, predicting and overall controlling nuclear<br>\n(power) plant (N(P)P) operation. Since then, the resulting CI-based implementations have afforded<br>\nincreasingly reliable as well as robust performance, demonstrating their potential as either stand-alone<br>\ntools, or - whenever more advantageous - combined with each other as well as with traditional signal<br>\nprocessing techniques. The present review is focused upon the application of CI methodologies to<br>\nthe - generally acknowledged as - key-issues of N(P)P operation, namely: control, diagnostics and<br>\nfault detection, monitoring, N(P)P operations, proliferation and resistance applications, sensor and<br>\ncomponent reliability, spectroscopy, fusion supporting operations, as these have been reported in the<br>\nrelevant primary literature for the period 1990-2015. At one end, 1990 constitutes the beginning of<br>\nthe actual implementation of innovative, and &ndash; at the same time &ndash; robust as well as practical, directly<br>\nimplementable in H/W, CI-based solutions/tools which have proved to be significantly superior to<br>\nthe traditional as well as the artificial-intelligence-(AI)derived methodologies in terms of operation<br>\nefficiency as well as robustness-to-noise and/or otherwise distorted/missing information. At the<br>\nother end, 2015 marks a paradigm shift in terms of the emergent (and, swiftly, ubiquitous) use of<br>\ndeep neural networks (DNNs) over existing ANN architectures and FL problem representations, thus<br>\ndovetailing the increasing requirements of the era of complex - as well as Big - Data and forever<br>\nchanging the means of ANN/neuro-fuzzy construction and application/performance. By exposing<br>\nthe prevalent CI-based tools for each key-issue of N(P)P operation, overall as well as over time for<br>\nthe given 1990-2015 period, the applicability and optimal use of CI tools to NE problems is revealed,<br>\nthus providing the necessary know-how concerning crucial decisions that need to be made for the<br>\nincreasingly efficient as well as safe exploitation of NE.</p>", 
  "author": [
    {
      "family": "Tatiana Tambouratzis T., Giannatsis J., Kyriazis A., and Siotropos P."
    }
  ], 
  "page": "27-109", 
  "volume": "9", 
  "type": "article-journal", 
  "issue": "1", 
  "id": "3579177"
}
24
39
views
downloads
Views 24
Downloads 39
Data volume 336.2 MB
Unique views 23
Unique downloads 36

Share

Cite as