Nonlinear model-based condition monitoring of advanced gas-cooled nuclear reactor cores

The graphite core is one critical component in gascooled nuclear reactors and it ages and degrades over time. As a result, the graphite core can dictate the life-time of a reactor in a nuclear power station. To support the safety cases and ensure the continued safe operation of an advanced gas-cooled reactor (AGR) nuclear plant, it is important to closely monitor the condition of its reactor graphite core to maintain the integrity throughout the life of the reactor. Toward this end, the fuel grab load trace (FGLT) measurements are currently used as main information sources to infer the core condition. Due to the fact that the FGLTs are masked by many effects in the refuelling process, the first principles models for nuclear refuelling process are promising to separate the information of interests to core condition from the masked FGLT measurements. To reliably and accurately obtain the unknown parameters existing in the developed first principles model for model-based condition monitoring of AGR nuclear graphite cores, this paper presents a nonlinear system identification approach. In this approach, a nonlinear first principles model is first developed to describe the refuelling process. A friction model is then investigated to mathematically deal with the frictional effects. The aerodynamic force is also modelled separately. Finally, the Trust- Region Reflective Newton method is used to find the optimal parameters in the nonlinear refuelling model. The real-world data from an AGR nuclear power plant is employed to demonstrate the effectiveness of the proposed nonlinear system identification approach for nonlinear model-based condition monitoring of graphite cores

[1]  Pascal Bigras,et al.  Parameter estimation for the LuGre friction model using interval analysis and set inversion , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[2]  Heinz Unbehauen,et al.  A review of identification in continuous-time systems , 1998 .

[3]  F. Allgöwer,et al.  Parameter and Derivative Estimation for Nonlinear Continuous-Time System Identification , 2001 .

[4]  Erfu Yang,et al.  Model-based condition monitoring of AGR nuclear graphite cores , 2010 .

[5]  Barry Marsden,et al.  Failure predictions for nuclear graphite using a continuum damage mechanics model , 2004 .

[6]  J. R. McDonald,et al.  An intelligent system for interpreting the nuclear refuelling process within an advanced gas-cooled reactor , 2003 .

[7]  J.R. McDonald,et al.  The Use of Hidden Markov Models for Anomaly Detection in Nuclear Core Condition Monitoring , 2009, IEEE Transactions on Nuclear Science.

[8]  Michael J. Grimble,et al.  Condition monitoring of the nuclear graphite core using benchmarking techniques , 2007, 2007 European Control Conference (ECC).

[9]  Michael J. Grimble,et al.  State and parameter estimation approach to monitoring AGR nuclear core , 2007 .

[10]  J.R. McDonald,et al.  Data mining reactor fuel grab load trace data to support nuclear core condition monitoring , 2006, IEEE Transactions on Nuclear Science.

[11]  Jan Swevers,et al.  EXPERIMENTAL COMPARISON OF DIFFERENT FRICTION MODELS FOR ACCURATE LOW-VELOCITY TRACKING , 2002 .

[12]  Thomas F. Coleman,et al.  A Subspace, Interior, and Conjugate Gradient Method for Large-Scale Bound-Constrained Minimization Problems , 1999, SIAM J. Sci. Comput..

[13]  Jorge J. Moré,et al.  Computing a Trust Region Step , 1983 .

[14]  Richard H. Byrd,et al.  Approximate solution of the trust region problem by minimization over two-dimensional subspaces , 1988, Math. Program..

[15]  C. Canudas-de-Wit.  Comments on "A new model for control of systems with friction" , 1998, IEEE Trans. Autom. Control..

[16]  P. Dahl.  Solid friction damping of spacecraft oscillations , 1975 .

[17]  Claudio Bonivento,et al.  Monitoring a gas-cooled nuclear reactor core integrity , 2008 .