Inteligența computațională aplicată sistemelor electronice utilizate în inginerie nucleară / Computational intelligence applied to electronic systems used in nuclear engineering

Computational intelligence in nuclear engineering has changed substantially in recent years. The main activities have been focused on the development of monitoring conditions and diagnostic techniques for nuclear power plants, but recent activities focus on the implementation and development of methods for the next generation of plants and reactors. One of the main purposes of computational intelligence is to understand the principles that make intelligent behavior possible in natural or artificial systems. In the case of engineering, the main purpose is to specify methods for designing intelligent and useful objects. The role of Computational Intelligence (CI) in the nuclear industry is in a constant transition due to operations, operational objectives, future needs and the requirements of a nuclear power plant. Many of the techniques developed in the last 30 years have reached maturity and new techniques are being applied. One of these is the non-destructive analysis with eddy currents of a fuel element. Other techniques such as non-destructive gamma-scanning and non-destructive visual analysis have reached near maturity in their development, being proposed for large-scale implementation. New techniques such as autonomous control and multi-intelligent agents are in the form of training, which are in the development cycle. Several papers on “state of the art” or current trends in the application of instrumentation and regulation (CI) techniques in nuclear power plants have been published on this topic. Applications have been developed on various electronic devices and / or platforms. Although the design and especially the optimization process require long periods of time, it is convenient to use algorithms based on fuzzy logic or neural networks for automatic optimization.