https://scholarworks.unist.ac.kr/handle/201301/67 2026-04-19T14:10:15Z https://scholarworks.unist.ac.kr/handle/201301/91339 Title: Sensor fault detection and signal restoration using scenario-guided supervised autoencoders for nuclear power plant accidents Author(s): Choi, Jeonghun; Koo, Seo Ryong; Lee, Seung Jun Abstract: Accurate detection and timely restoration of faulty sensor signals are critical for ensuring safety and operational integrity in nuclear power plants, especially under accident conditions characterized by highly dynamic and nonlinear parameter behaviors. In this study, we propose a scenario-guided supervised autoencoder (SAE) framework that performs three integrated functions: (1) detecting faulty sensor signals during accident transients, (2) restoring corrupted signals to their unfaulted trajectories, and (3) preserving the quality of downstream accident diagnosis. Our framework employs a supervised variational autoencoder (VAE) with a long shortterm memory encoder, trained on accident scenario simulations from a compact nuclear simulator replicating key thermal-hydraulic behaviors of a pressurized water reactor. A classification decoder provides scenario-guided supervision, enabling the model to distinguish between fault-induced signal deviations and legitimate accident transient changes. Evaluation across nine accident scenarios and seven different sensor fault types shows that the proposed SAE achieves a fault detection true positive rate of 99.09% with 95.52% precision, compared to 95.50% and 91.99% for the conventional VAE. For signal restoration, 97.70% of restored signals fall within 15% mean absolute error of their unfaulted trajectories, compared to 69.50% for the VAE. The restoration-based approach recovers accident diagnosis accuracy to levels comparable to unfaulted conditions, outperforming fault isolation strategies for most fault types. Additional robustness analyses show that the proposed framework retains detection performance under additive Gaussian measurement noise up to sigma = 0.05 (5% of the normalized signal range) and achieves a 100% fault detection rate for up to three simultaneous sensor faults. These results suggest that scenario-guided supervised autoencoders can improve sensor signal integrity in safety-critical nuclear applications. 2026-05-31T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/91248 Title: Electrorefining of HANA-4 Cladding Scrap in LiCl-KCl Salts for Volumetric Decontamination of Irradiated Cladding Containing Nb-94 Author(s): Son, Sungjune; Hur, Jungho; Park, Jaeyeong; Kim, Pyeong-Hwa; Hwang, Il Soon Abstract: Nb-94, an activation product in irradiated cladding, is a major concern in geological disposal. As the Nb-94 is distributed throughout the cladding, volumetric decontamination should be applied to separate between Zr and Nb. In this study, the radiological characteristics of irradiated HANA-4 cladding were investigated using the ORIGEN-ARP code to derive a decontamination factor for Nb-94, 11. The electrochemical behavior of Nb was evaluated by cyclic voltammetry using low NbCl5 concentrations (0.15 and 0.5 wt. %) in LiCl-KCl at 773 K. Nb has a complex redox behavior but it was found the nobler tendency could be utilized for Zr electrorefining. Two electrorefining tests were performed by applying constant potentials of -0.85 V and -1.2 V (vs. Ag/AgCl 1 wt. %) at the anode and cathode, respectively. From the anodic test, Zr metal was obtained at the bottom of the salt, by two-step reactions among Zr4+ ,Zr2+, and Zr. Meanwhile, ZrCl and Zr were co-recovered as deposits in the cathodic test. The results revealed good separation performance between Zr and Nb. The Nb concentrations were 2.1 and 20.3 ppm in the product from the anodic and cathodic test, respectively, supporting feasibility of satisfying the radioactivity concentration limits of Gyeongju Disposal Facility. 2026-06-30T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/91172 Title: Development of GPU-Accelerated CMFD for STREAM3D-GPU Neutron Transport Code Author(s): Setiawan, Fathurrahman; Dzianisau, Siarhei; Lee, Deokjung Abstract: This study presents the implementation of the graphics processing unit (GPU)-based coarse mesh finite difference (CMFD) acceleration in STREAM3D-GPU using the directive-based OpenACC framework. The offloading process follows a structured approach of assessment, parallelization, and optimization, with data structures reorganized to maximize GPU efficiency. Performance evaluations on three-dimensional OPR-1000 reactor models demonstrate up to a 22-fold reduction in CMFD run time compared to the CPU version, with the greatest improvements observed in the linear system solver and flux convergence routines due to extensive parallelization and concurrent execution. Numerical verification using depletion simulations of the BEAVRS benchmark confirmed that the GPU implementation maintained high fidelity, with eigenvalue deviations within 5 pcm and maximum differences in power distribution below 0.6%. 2026-02-28T15:00:00Z https://scholarworks.unist.ac.kr/handle/201301/91036 Title: Radiation safety assessment of a worker for the treatment facilities of the mixed spent resin from heavy water reactor Author(s): Yoon, Ja Yeong Abstract: The treatment facility for mixed spent resin at Wolsong CANDU nuclear power plant involves potential radiation exposure risks to workers during operation. This study established realistic normal and accident scenarios in advance to ensure long term and stable operation of the treatment facility to be installed, and provided the number of operations, time, and methods for conducting a radiation safety assessment. The primary components of the treatment facility include 1 kW and 9 kW scale beta-nuclide removal unit. This research evaluated shielding safety of the beta-nuclide removal units, analyzed worker doses during waste transport before and after treatment and examined exposure during injection, assessed doses under the leakage scenario of radioactive airborne particulates and maintenance operations. The study established realistic operational scenarios through empirical experiments and quantitatively analyzed both external and internal dose rates of workers. Using the VISIPLAN 4.0 code, the external dose rates of radionuclides contained in the mixed spent resin were evaluated, and internal dose rates were calculated considering the facility volume, breathing rate, and working time. For the shielding safety evaluation, lead and water shields of varying thickness were applied, and increasing shielding thickness consistently reduced spatial dose rates. The before treatment scenario assumed normal operation without leakage, while the after treatment scenario evaluated worker doses by accounting for the production of 5 kg of spent resin and 0.5 kg of combined zeolite and activated carbon. The injection scenario involved lifting the reactor lid and loading 2 kg of mixed spent resin into the facility. The leakage scenario of radioactive airborne particulates, the analysis considered a 20 % generation of radioactive airborne particulates relative to the unit volume for both the 1 kW and 9 kW scale beta-nuclide removal unit. And the maintenance scenario simulated pipe replacement based on an actual blockage incident. Therefore, this study conservatively assessed radiation safety by presenting work methods based on an annual dose of 20 mSv. Mixed spent resin treatment facility is currently in the pre-installation stage, during which various feasibility and performance verifications are being conducted. Accordingly, presenting measures to maintain radiological protection for workers during future operation will provide essential baseline data for securing the long-term safe operation of the facility. Major: Department of Nuclear Engineering 2026-01-31T15:00:00Z