A novel neural/symbolic hybrid approach to heuristically optimized fuel allocation and automated revision of heuristics in nuclear engineering

FUELCON is an expert system for optimized refueling design in nuclear engineering. This task is crucial for keeping down operating costs at a plant without compromising safety. FUELCON proposes sets of alternative configurations of allocation of fuel assemblies that are each positioned in the planar grid of a horizontal section of a reactor core. Results are simulated, and an expert user can also use FUELCON to revise rulesets and improve on his or her heuristics. The successful completion of FUELCON led this research team into undertaking a panoply of sequel projects, of which we provide a meta-architectural comparative formal discussion. In this paper, we demonstrate a novel adaptive technique that learns the optimal allocation heuristic for the various cores. The algorithm is a hybrid of a fine-grained neural network and symbolic computation components. This hybrid architecture is sensitive enough to learn the particular characteristics of the ‘in-core fuel management problem’ at hand, and is powerful enough to use this information fully to automatically revise heuristics, thus improving upon those provided by a human expert.

Refueling in nuclear engineering: the FUELCON project

Over a time span of almost a decade, the FUELCON project in nuclear engineering has led to a fully functional expert system and spawned sequel projects. Its task is in-core fuel management, also called `refueling', i.e., good fuel-allocation for reloading the core of a given nuclear reactor, for a given operation cycle. The task is crucial for keeping down operation costs at nuclear power plants. Fuel comes in different types and is positioned in a grid representing the core of a reactor. The tool is useful for practitioners but also helps the expert in the domain to test his or her rules of thumb and to discover new ones.

Upgrading automation for nuclear fuel in-core management: from the symbolic generation of configurations to the neural adaptation of heuristics

FUELCON is an expert system in nuclear engineering. Its task is optimized refueling-design, which is crucial to keep down operation costs at a plant. FUELCON proposes sets of alternative configurations of fuel-allocation; the fuel is positioned in a grid representing the core of a reactor. The practitioner of in-core fuel management uses FUELCON to generate a reasonably good configuration for the situation at hand. The domain expert, on the other hand, resorts to the system to test heuristics and discover new ones, for the task described above. Expert use involves a manual phase of revising the ruleset, based on performance during previous iterations in the same session. This paper is concerned with a new phase: the design of a neural component to carry out the revision automatically. Such an automated revision considers previous performance of the system and uses it for adaptation and learning better rules. The neural component is based on a particular schema for a symbolic to recurrent-analogue bridge, called NIPPL, and on the reinforcement learning of neural networks for the adaptation.

Reinforcement corrosion initiation and activation times in concrete structures exposed to severe marine environments

The corrosion of steel reinforcement bars in reinforced concrete structures exposed to severe marine environments usually is attributed to the aggressive nature of chloride ions. In some cases in practice corrosion has been observed to commence already within a few years of exposure even with considerable concrete cover to the reinforcement and apparently high quality concretes. However, there are a number of other cases in practice for which corrosion initiation took much longer, even in cases with quite modest concrete cover and modest concrete quality. Many of these structures show satisfactory long-term structural performance, despite having high levels of localized chloride concentrations at the reinforcement. This disparity was noted already more than 50 years ago, but appears still not fully explained. This paper presents a systematic overview of cases reported in the engineering and corrosion literature and considers possible reasons for these differences. Consistent with observations by others, the data show that concretes made from blast furnace cements have better corrosion durability properties. The data also strongly suggest that concretes made with limestone or non-reactive dolomite aggregates or sufficiently high levels of other forms of calcium carbonates have favourable reinforcement corrosion properties. Both corrosion initiation and the onset of significant damage are delayed. Some possible reasons for this are explored briefly.