On-line neuro-expert monitoring system for Borssele Nuclear Power Plant

A new method for an on-line monitoring system for the nuclear power plants has been developed utilizing the neural networks and the expert system. The integration of them is expected to enhance a substantial potential of the functionality as operators support.

The recurrent neural network and the feed-forward neural network with adaptive learning are selected for the plant modeling and anomaly detection because of the high capability of modeling for dynamic behavior. The expert system is used as a decision agent, which works on the information space of both the neural networks and the human operators. The information of other sensory signals is also fed to the expert system, together with the outputs that the neural networks generate from the measured plant signals. The expert system can treat almost all known correlation between plant status patterns and operation modes as a priori set of rules.

From the off-line test at Borssele Nuclear Power Plant (PWR 480 MWe) in the Netherlands, it was shown that the neuro-expert system successfully monitored the plant status. The expert system worked satisfactorily in diagnosing the system status by using the outputs of the neural networks and a priori knowledge base from the PWR simulator. The electric power coefficient is simultaneously monitored from the measured reactive and active electric power signals.     

First-Principles Study on the Grain Boundary Embrittlement of Metals by Solute Segregation: Part II. Metal (Fe,Al, Cu)-Hydrogen (H) Systems

The microscopic mechanism of grain boundary (GB) embrittlement in metals by hydrogen segregation (trapping) has been not well understood for many years. From first-principles calculations, we show here that the calculated cohesive energy of bcc Fe Σ3(111) and fcc Al(Cu) Σ5(012) symmetrical tilt GBs can be significantly reduced if many hydrogen atoms segregate at the GBs. This indicates that the reduction of the cohesive energy of the GB may cause the hydrogen-induced GB embrittlement in Fe, Al, and Cu. Considering the “mobile” effect of hydrogen during fracture, especially for the Fe system, more hydrogen atoms coming from solid solution state can segregate on the gradually formed two fracture surfaces and reduce further the cohesive energy. We suggest a new idea about the upper and lower critical stresses observed in the constant-load test of hydrogen-induced delayed fracture in high-strength steels; the upper critical stress is determined by the amount (density) of “immobile” hydrogen atoms segregated at the GB before fracture, and the lower critical stress is determined by the total amount (density) of immobile and mobile hydrogen atoms, the latter of which segregate on the two fracture surfaces during fracture.     

Application of a nonlinear dynamical descriptor to BWR stability analyses

Reactor noise analyis method based on nonlinear dynamical theory is applied to the Forsmarks 1&2 BWR stability benchmark organized by OECD/NEA. The method utilizes the determination process of the fractal dimension of oscillatory neutron-flux signals. For practical application, the fractal dimension is expected to classify different asymptotic regimes of nonlinear dynamical systems. In this case, each signal is classified into stable, quasi-stable, and unstable states. It was confirmed that the result was consistent to that of decay ratios. In addition, because the data processing does not include fitting calculation often used by the decay ratio, it is surmised that the result hardly varies with analyzers. This can be the prominent advantage of this methodology compared to decay ratio.     

Software integration for monitoring systems with high flexibility

A new scheme of software integration methodology is applied to the implementation of a neural-network-based real-time monitoring aid system. In this scheme, the data communication between modules is established by means of connecting a standard I/O stream of a module to that of another. The methodology enables easy coupling of software modules with the least modifications of existing source code, and it can make distributed software systems highly flexible, portable and testable.