Online sequential condition prediction method of natural circulation systems based on EOS-ELM and phase space reconstruction

Abstract Natural circulation design is widely used in the passive safety systems of advanced nuclear power reactors. The irregular and chaotic flow oscillations are often observed in boiling natural circulation systems so it is difficult for operators to monitor and predict the condition of these systems. An online condition forecasting method for natural circulation system is proposed in this study as an assisting technique for plant operators. The proposed prediction approach was developed based on Ensemble of Online Sequential Extreme Learning Machine (EOS-ELM) and phase space reconstruction. Online Sequential Extreme Learning Machine (OS-ELM) is an online sequential learning neural network algorithm and EOS-ELM is the ensemble method of it. The proposed condition prediction method can be initiated by a small chunk of monitoring data and it can be updated by newly arrived data at very fast speed during the online prediction. Simulation experiments were conducted on the data of two natural circulation loops to validate the performance of the proposed method. The simulation results show that the proposed predication model can successfully recognize different types of flow oscillations and accurately forecast the trend of monitored plant variables. The influence of the number of hidden nodes and neural network inputs on prediction performance was studied and the proposed model can achieve good accuracy in a wide parameter range. Moreover, the comparison results show that the proposed condition prediction method has much faster online learning speed and better prediction accuracy than conventional neural network model.

[1]  Kenji Fukuda,et al.  Analysis of the critical heat flux in round vertical tubes under low pressure and flow oscillation conditions. Applications of artificial neural network , 2003 .

[2]  Emine Ayaz.  Component-wide and plant-wide monitoring by neural networks for Borssele nuclear power plant , 2008 .

[3]  T. L. Schulz,et al.  Westinghouse AP1000 advanced passive plant , 2006 .

[4]  Manmohan Pandey,et al.  Stability analysis and nonlinear dynamics of natural circulation boiling water reactors , 2008 .

[5]  S. Jiang,et al.  Experimental simulation study on start-up of the 5 MW nuclear heating reactor , 1995 .

[6]  Narasimhan Sundararajan,et al.  A Fast and Accurate Online Sequential Learning Algorithm for Feedforward Networks , 2006, IEEE Transactions on Neural Networks.

[7]  K. Nabeshima,et al.  Nuclear reactor monitoring with the combination of neural network and expert system , 2002, Math. Comput. Simul..

[8]  Manmohan Pandey,et al.  Parametric effects on reactivity instabilities and nonlinear oscillations in a nuclear-coupled double channel natural circulation boiling system , 2010 .

[9]  Guanghui Su,et al.  Experimental study on two-phase flow instability of natural circulation under rolling motion condition , 2009 .

[10]  S. Y. Jiang,et al.  Experimental Study of Two-Phase Flow Oscillation in Natural Circulation , 2000 .

[11]  A. Massi Pavan,et al.  On-line fault detection of a fuel rod temperature measurement sensor in a nuclear reactor core using ANNs , 2015 .

[12]  Guang-Bin Huang,et al.  Extreme learning machine: a new learning scheme of feedforward neural networks , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[13]  Fei Xie,et al.  Prediction of time series of NPP operating parameters using dynamic model based on BP neural network , 2015 .

[14]  Sichao Tan,et al.  Non-linear time series analysis on flow instability of natural circulation under rolling motion condition , 2014 .

[15]  R. Penrose.  A Generalized inverse for matrices , 1955 .

[17]  James P. Crutchfield,et al.  Geometry from a Time Series , 1980 .

[18]  Enrico Zio,et al.  Nuclear Power Plant Components Condition Monitoring by Probabilistic Support Vector Machine , 2013 .

[19]  Ji Xing,et al.  HPR1000: Advanced Pressurized Water Reactor with Active and Passive Safety , 2016 .

[20]  Narasimhan Sundararajan,et al.  Fully complex extreme learning machine , 2005, Neurocomputing.

[21]  Zhong-ning Sun,et al.  An investigation on flashing instability induced water hammer in an open natural circulation system , 2016 .

[22]  Narasimhan Sundararajan,et al.  On-Line Sequential Extreme Learning Machine , 2005, Computational Intelligence.

[23]  Chung-Yun Wu,et al.  Chaotic oscillations in a low pressure two-phase natural circulation loop under low power and high inlet subcooling conditions , 1996 .

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

[25]  Emine Ayaz,et al.  Comparisons between the various types of neural networks with the data of wide range operational conditions of the Borssele NPP , 2003 .

[26]  Yuan Lan,et al.  Ensemble of online sequential extreme learning machine , 2009, Neurocomputing.

[27]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.