Event Identification in Prototype Fast Breeder Reactor Subsystem using Artificial Neural Network

Identification of events in nuclear power plant is a very challenging task because each event has a unique set of patterns based on the dynamic behavior of the plant. Accordingly the effective set of parameters has to be chosen to identify an event having a particular pattern. This paper describes the development of artificial neural network (ANN) model for event identification of Primary Sodium System of Prototype Fast Breeder Reactor. In reactor under normal operating condition, the Primary Sodium Pump takes the sodium from the cold pool towards the core. Due to some mechanical and electrical problems the Primary Sodium Pump trip and Primary Sodium Pump Seizure may occur which can be identified manually after analyzing the process parameters. The work involves implementation of two ANN models to identify the occurrence of these two events. The effective parameters considered for these two events are SCRAM (Safety Control Rod Accelerated Movement) parameters. The training data for modeling the neural network is prepared using the thermo hydraulics simulation code of PFBR simulator. Multilayer neural network using back propagation algorithm has been widely used for transient identification. The proposed ANN model was able to identify the events correctly and the results obtained are satisfactory. General Terms Back Propagation Algorithms, Event Identification

[1]  A. Srivastava,et al.  Diagnostic system for identification of accident scenarios in nuclear power plants using artificial neural networks , 2009, Reliab. Eng. Syst. Saf..

[2]  Enrico Zio,et al.  A data-driven approach for predicting failure scenarios in nuclear systems , 2010 .

[3]  B. Yegnanarayana,et al.  Artificial Neural Networks , 2004 .

[4]  Timo Sorsa,et al.  Neural networks in process fault diagnosis , 1991, IEEE Trans. Syst. Man Cybern..

[5]  Poong Hyun Seong,et al.  A dynamic neural network aggregation model for transient diagnosis in nuclear power plants , 2007 .

[6]  S. B. Bhoje,et al.  DESIGN OF 500 MWe PROTOTYPE FAST BREEDER REACTOR , 2004 .

[7]  Gopika Vinod,et al.  Simulation of Pressurised Heavy Water Reactor Data Using Artificial Neural Network for Reactor Status / Transient Identification , 2002 .

[8]  Kamal Hadad,et al.  Fault diagnosis and classification based on wavelet transform and neural network , 2011 .

[9]  J. W. Hines,et al.  Pattern Recognition Techniques for Transient Detection to Enhance Nuclear Reactors’ Operational Safety , 2000 .

[10]  Mark J. Embrechts,et al.  Hybrid identification of nuclear power plant transients with artificial neural networks , 2004, IEEE Transactions on Industrial Electronics.

[11]  Davide Roverso.  Neural Ensembles for Event Identification , 2000 .

[12]  S. Sumathi,et al.  Introduction to neural networks using MATLAB 6.0 , 2006 .

[13]  Atul Srivastava,et al.  A diagnostic system for identifying accident conditions in a nuclear reactor , 2011 .

[14]  A.G. Parlos,et al.  Artificial neural networks based system identification and control of nuclear power plant components , 1990, 29th IEEE Conference on Decision and Control.

[15]  Csilla Bányász,et al.  Fault detection, supervision, and safety for technical processes 2000 (SAFEPROCESS 2000) : a proceedings volume from the 4th IFAC symposium, Budapest, Hungary, 14-16 June, 2000 , 2001 .

[16]  Jin Jiang,et al.  Applications of fault detection and diagnosis methods in nuclear power plants: A review , 2011 .

[17]  S. N. Sivanandam,et al.  Principles of soft computing , 2011 .