The amount of radioactive waste is expected to dramatically increase with decommissioningof nuclear power plants such as Kori-1, the first nuclear power plant in South Korea. Accuratenuclide analysis is necessary to manage the radioactive wastes safel...
The amount of radioactive waste is expected to dramatically increase with decommissioningof nuclear power plants such as Kori-1, the first nuclear power plant in South Korea. Accuratenuclide analysis is necessary to manage the radioactive wastes safely, but research on verification ofradionuclide analysis has yet to be well established. This study aimed to develop the technology thatcan verify the results of radionuclide analysis based on artificial intelligence. In this study, we proposean anomaly detection algorithm for inspecting the analysis error of radionuclide. We used the datafrom ‘Updated Scaling Factors in Low-Level Radwaste’ (NP-5077) published by EPRI (Electric PowerResearch Institute), and resampling was performed using SMOTE (Synthetic Minority OversamplingTechnique) algorithm to augment data. 149,676 augmented data with SMOTE algorithm was used totrain the artificial neural networks (classification and anomaly detection networks). 324 NP-5077 reportdata verified the performance of networks. The anomaly detection algorithm of radionuclide analysiswas divided into two modules that detect a case where radioactive waste was incorrectly classified ordiscriminate an abnormal data such as loss of data or incorrectly written data. The classification networkwas constructed using the fully connected layer, and the anomaly detection network was composedof the encoder and decoder. The latter was operated by loading the latent vector from the end layer ofthe classification network. This study conducted exploratory data analysis (i.e., statistics, histogram, correlation, covariance, PCA, k-mean clustering, DBSCAN). As a result of analyzing the data, it iscomplicated to distinguish the type of radioactive waste because data distribution overlapped each other.
In spite of these complexities, our algorithm based on deep learning can distinguish abnormal data fromnormal data. Radionuclide analysis was verified using our anomaly detection algorithm, and meaningfulresults were obtained.