Muhammad Subekti, MS (2009) THE IMPROVEMENT OF NEURO-EXPERT METHOD FOR ANOMALY DETECTION IN NUCLEAR REACTOR. THE IMPROVEMENT OF NEURO-EXPERT METHOD FOR ANOMALY DETECTION IN NUCLEAR REACTOR. pp. 222-230. ISSN 0854-2910
25_M. Subekti_222-230_OK.pdf
Download (361kB) | Preview
Abstract
ABSTRACT
The improvement of Neuro-Expert Method have been done for anomaly detection in nuclear reactor by
utilization of Recurrent Neural Network (RNN). The development of Multilayer Perceptron (MLP) for similar
objective was done in the previous research. Due to the monitoring system needs of redundancy to assure the
reactor safety, the Neuro-Expert have been improved in this research by utilization of RNN as added method.
Furthermore, the expert system is coupled to MLP and RNN by using specific parameters which depend on
the training characteristic. The offline demonstation of MLP and RNN were carry out for PWR Borselle,
simulator of PWR Surry-1, RSG-GAS reactor, and High Temperature Engineering Tested Reactor (HTTR).
The learning results showed unsignificant different of maximum error of 0.0061 for MLP and 0.0049 for
Jordan typed RNN. In the contrary, Elman typed RNN gave unreliable maximum error of 0.0130.
Keywords: improvement, Neuro-Expert, multilayer perceptron, Recurrent Neural Network, anomaly
detection.
ABSTRAK
Perbaikan metode Neuro-Expert untuk mendeteksi anomali di reaktor nuklir dengan memanfaatkan
Recurrent Neural Network (RNN) telah dilakukan. Pada penelitian sebelumnya, pengembangan metode
Multilayer Perceptron (MLP) untuk tujuan yang sama telah dilakukan. Karena sistem monitoring
memerlukan redundansi untuk menjamin keselamatan reaktor, maka pada penelitian sekarang ini, metode
Neuro-Expert diperbaiki dengan memanfaatkan RNN sebagai metode tambahan. Dan selanjutnya, sistem
pakar pada MLP dan RNN menggunakan parameter optimasi yang spesifik sesuai dengan karakteristik
pelatihan. Demonstrasi MLP dan RNN secara offline telah dilakukan dengan menggunakan data PWR
Borselle, simulator PWR Surry-1, Reaktor RSG-GAS, dan High Temperature Engineering Tested Reactor
(HTTR). Hasil pelatihan menunjukkan perbedaan yang tidak signifikan dimana maximum error sebesar
0,0061 untuk MLP dan 0,0049 untuk RNN tipe Jordan. Sebaliknya, RNN tipe Elman memberikan maximum
error sebesar 0,0130.
Katakunci: improvement, Neuro-Expert, multilayer perceptron, Recurrent Neural Network, anomaly
detection.
Item Type: | Article |
---|---|
Subjects: | Taksonomi BATAN > Keselamatan dan Keamanan Nuklir Taksonomi BATAN > Keselamatan dan Keamanan Nuklir |
Divisions: | BATAN > Pusat Teknologi dan Keselamatan Reaktor Nuklir IPTEK > BATAN > Pusat Teknologi dan Keselamatan Reaktor Nuklir |
Depositing User: | Administrator Repository |
Date Deposited: | 12 Nov 2018 03:24 |
Last Modified: | 31 May 2022 03:33 |
URI: | https://karya.brin.go.id/id/eprint/4891 |