Prediction of Power Signal in Nuclear Reactors with Neural Network Based Intelligent Predictors in the Presence of 1/fα type Sensor Noise

被引：3

作者：

Majumder, Basudev ^{[1
]}

Saha, Sayan ^{[2
]}

Das, Saptarshi ^{[1
,3
]}

Pan, Indranil ^{[3
]}

Gupta, Amitava ^{[1
,3
]}

机构：

[1] Jadavpur Univ, Sch Nucl Studies & Applicat SNSA, Sect 3, Salt Lake Campus,LB 8, Kolkata 700098, India

[2] Jadavpur Univ, Dept Instrumentat & Elect Engn, Sect 3, Kolkata 700098, India

[3] Jadavpur Univ, Dept Power Engn, Sect 3, Kolkata 700098, India

来源：

MEMS, NANO AND SMART SYSTEMS, PTS 1-6 | 2012年 / 403-408卷

关键词：

ANN; dynamic neural network; fractional order noise; prediction; RBF network; reactor power; PLANT; IDENTIFICATION; MODEL;

D O I：

10.4028/www.scientific.net/AMR.403-408.4512

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper describes the prediction of the power output from the Self-Powered-Neutron-Detector (SPND) in the liquid zone control compartment (LZC) of nuclear reactors, which are important in online global power measurement in a large Pressurized Heavy Water Reactor (PHWR). Noisy measured data from the SPNDs have been smoothened out with the help of an ARMA filter and then the smoothed data is used as the input for the neural networks for training purpose. These typical intelligent predictors have been studied with its variation considering different dynamic neural network structures with integer and fractional order noise considerations in the measured sensor data. The paper reports the best found network structure for the prediction of measured set of noisy SPND data.

引用

页码：4512 / +

页数：4

共 24 条

[1] Application of the Wiener-Hermite Functional method to a linear feedback model of point reactor kinetics driven by random reactivity noise [J].

Behringer, K ;

Pineyro, J .

ANNALS OF NUCLEAR ENERGY, 1997, 24 (08) :587-623

[2] Identification of a nuclear reactor core (VVER) using recurrent neural networks [J].

Boroushaki, M ;

Ghofrani, MB ;

Lucas, C .

ANNALS OF NUCLEAR ENERGY, 2002, 29 (10) :1225-1240

[3] Simulating the dynamics of the neutron flux in a nuclear reactor by locally recurrent neural networks [J].

Cadini, F. ;

Zio, E. ;

Pedroni, N. .

ANNALS OF NUCLEAR ENERGY, 2007, 34 (06) :483-495

[4] Computational intelligence-based trajectory scheduling for control of nuclear research reactors [J].

Coban, Ramazan .

PROGRESS IN NUCLEAR ENERGY, 2010, 52 (04) :415-424

[5] Numerical tools applied to power reactor noise analysis [J].

Demaziere, Christophe ;

Pazsit, Imre .

PROGRESS IN NUCLEAR ENERGY, 2009, 51 (01) :67-81

[6] Estimation of research reactor core parameters using cascade feed forward artificial neural networks [J].

Hedayat, Afshin ;

Davilu, Hadi ;

Barfrosh, Ahmad Abdollahzadeh ;

Sepanloo, Kamran .

PROGRESS IN NUCLEAR ENERGY, 2009, 51 (6-7) :709-718

[7] DISCRETE SIMULATION OF COLORED NOISE AND STOCHASTIC-PROCESSES AND 1/F(ALPHA) POWER-LAW NOISE GENERATION [J].

KASDIN, NJ .

PROCEEDINGS OF THE IEEE, 1995, 83 (05) :802-827

[8] APPLICATION OF NEURAL NETWORKS TO SIGNAL PREDICTION IN NUCLEAR-POWER-PLANT [J].

KIM, WJ ;

CHANG, SH ;

LEE, BH .

IEEE TRANSACTIONS ON NUCLEAR SCIENCE, 1993, 40 (05) :1337-1341

[9] Design and optimization of fuzzy-PID controller for the nuclear reactor power control [J].

Liu, Cheng ;

Peng, Jin-Feng ;

Zhao, Fu-Yu ;

Li, Chong .

NUCLEAR ENGINEERING AND DESIGN, 2009, 239 (11) :2311-2316

[10] Predicting neutron diffusion eigenvalues with a query-based adaptive neural architecture [J].

Lysenko, MG ;

Wong, HI ;

Maldonado, GI .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1999, 10 (04) :790-800

← 1 2 3 →