Autonomous operation algorithm for safety systems of nuclear power plants by using long-short term memory and function-based hierarchical framework

Abstract With the improvement of computer performance and the emergence of cutting-edge artificial intelligence (AI) algorithms, an autonomous operation based on AI is being applied to many industries. An autonomous algorithm is a higher-level concept than conventional automatic operation in nuclear power plants (NPPs). In order to achieve autonomous operation, the autonomous algorithm needs to include superior functions to monitor, control and diagnose automated subsystems. This study suggests an autonomous operation algorithm for NPP safety systems using a function-based hierarchical framework (FHF) and a long short-term memory (LSTM). The FHF hierarchically models the safety goals, functions, systems, and components in the NPP. Then, the hierarchical structure is transformed into an LSTM network that is an evolutionary version of a recurrent neural network. This approach is applied to a reference NPP, a Westinghouse 930 MWe, three-loop pressurized water reactor. This LSTM network has been trained and validated using a compact nuclear simulator.

[1]  Panos J. Antsaklis,et al.  An introduction to autonomous control systems , 1991 .

[2]  Michael Y. Hu,et al.  Forecasting with artificial neural networks: The state of the art , 1997 .

[3]  Emine Ayaz,et al.  Elman's recurrent neural network applications to condition monitoring in nuclear power plant and rotating machinery , 2003 .

[4]  James A. Reggia,et al.  A generalized LSTM-like training algorithm for second-order recurrent neural networks , 2012, Neural Networks.

[5]  Gopika Vinod,et al.  Application of artificial neural networks to nuclear power plant transient diagnosis , 2007, Reliab. Eng. Syst. Saf..

[6]  Charles E. Billings,et al.  Aviation Automation: The Search for A Human-centered Approach , 1996 .

[7]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[8]  Pascal Vincent,et al.  Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives , 2012, ArXiv.

[9]  H. Basher.  Autonomous Control of Nuclear Power Plants , 2003 .

[10]  Frank L. Lewis,et al.  Guest Editorial Special Issue on New Developments in Neural Network Structures for Signal Processing, Autonomous Decision, and Adaptive Control , 2017, IEEE Trans. Neural Networks Learn. Syst..

[11]  O'hara,et al.  EXAMINING CONCEPT OF OPERATIONS IN FUTURE PLANTS. , 2004 .

[12]  Jürgen Schmidhuber,et al.  Learning to forget: continual prediction with LSTM , 1999 .

[13]  Herbert A. Simon,et al.  The Sciences of the Artificial , 1970 .

[14]  George F. Flanagan,et al.  Development of an Automated Decision-Making Tool for Supervisory Control System , 2014 .

[15]  Belle R. Upadhyaya,et al.  Autonomous Control of Space Reactor Systems , 2007 .

[16]  J. C. Higgins,et al.  Human-system Interfaces to Automatic Systems: Review Guidance and Technical Basis , 2010 .

[17]  Mohammad Javad Yazdanpanah,et al.  An intelligent nuclear reactor core controller for load following operations, using recurrent neural networks and fuzzy systems , 2003 .