Fuzzy identification of transients in nuclear power plants

Abstract The identification of transients is of fundamental importance for the timely monitoring of nuclear plants operation. The main target is detecting the occurrence of the onset of a transient, in its early stages, and identifying its kind so as to be able to readily act to fix its causes. Given the safety and economical importance of the problem, various approaches have been investigated and applied for transient identification, and many efforts are still devoted to the improvement of the results so far obtained. In this paper, a fuzzy-logic based method for the identification of transients is proposed. The method is ‘model-free’ in that the if-then rules, which constitute the heart of the approach, are inferred only from the available input-output signal data. The method is tested on an example of identification of reactor transients generated by four forcing functions of different nature. The necessary data for the identification have been simulated by the QUAndry-based Reactor Kinetics code (QUARK, distributed freely by NEA) configured so as to model the operations of the Westinghouse Advanced Pressurized water reactor, AP600.