Belief-based diagnosis module for turbine cycles in NPPs

Now the enhancement of efficiency is world-wide trend to survive under competition. In recent years, the national electricity reform performed in Korea made concern about turbine cycle higher in nuclear power plants (NPPs) as well as fossil-fuel plants. Authors have investigated actual situation to make a technical road-map for the efficiency enhancement of turbine cycle of NPPs. The final requirement of the technical road-map was the accurate identification of the root causes for electric loss. In this study, authors tried to make the model consisting of statistical regression and Bayesian network for quantitative diagnosis. Authors selected 34 boundary parameters at component level that independently influence to electric output. Using the data from a turbine cycle simulation tool, the statistical correlation between a boundary parameter and electric output was analyzed. For the identification or root causes under the uncertainty, belief module is added on the basis of Bayesian network for the boundary parameters. In conclusion, this diagnosis module can give the impact and the uncertainty of the root causes at the same time. This module was temporarily installed in Younggwang NPP unit 3 and 4 and is being under field testing. (author)