Intelligent systems integration: Guiding Principles, examples and lessons learned

The adoption of support systems based on computational intelligence (CI) techniques in nuclear power plants presents a number of challenges that need to be addressed in order to maximise the added value of this type of systems. This paper tackles issues related to the integration of CI systems and identifies three different levels of integration that should be taken into consideration. These are the data level, the operational level, and the functional level. The data level includes all aspects of data communication between the CI system, the plant instrumentation, and the other computerized support systems present at the plant. The operational level includes all those aspects of integration that influence the way in which the operator interacts with the CI system, with a clear emphasis on Human System Interface (HSI) integration. The functional level includes those aspects of integration that deal with the functionality of the CI system and how this can support or be combined with the functionality offered by other support systems with the aim of exploiting synergistic effects and achieving new functions. Additionally, the paper identifies a set of guiding principles of integration that can be applied at all three levels of integration. Examples taken from experiences gathered at the OECD Halden Reactor Project are also included.