Inferential Neural Networks for Nuclear Power Plant Sensor Channel Drift Monitoring

An artificial neural network (ANN) modeling technique is introduced for sensor and associated instrument channel validation. This method utilizes a genetic algorithm search approach supplemented with standard linear correlations to empirically determine appropriate combinations of available input variables to perform each desired modeling. The Sequential Probability Ratio Test (SPRT), a statistical decision technique, is investigated for compatibility and utility with ANN models. This drift detection system demonstrated with data supplied by Florida Power & Light Company’s Crystal River 3 Nuclear Power Plant.