Search Results (210)

Temperature sensor systems are critical for nuclear power plant (NPP) condition monitoring, whose reliability underpins unit safety and stability. Fault localization and diagnosis are essential to sustain their stable service. Conventional Principal Component Analysis (PCA) and Graph Neural Network (GNN) methods suffer clear drawbacks: PCA is vulnerable to noise and cannot classify fault types accurately, while GNNs struggle to quantify correlations among temperature data. This paper fuses PCA’s anomaly representation capability and GNN’s structural feature extraction capacity to propose an Abnormal Feature-GCN method for joint fault localization and diagnosis. First, an Adaptive PCA (APCA) model fed with multi-dimensional sensor features computes abnormal features. These features are then transformed into edge weights to construct a weighted graph. A dual-branch GCN is finally trained via a joint loss function for parallel multi-task learning to simultaneously locate faulty sensors and identify fault types. Validated on a nuclear primary circuit temperature sensor system under constant-, rising-, and falling-temperature working conditions, the proposed method realizes accurate fault localization and classification. The mean overall accuracy of the proposed method surpasses mainstream baselines by 2.23%, 1.62%, and 1.89% for the three typical working conditions, respectively. Full article

Operator diagnosis level in a main control room (MCR) of Nuclear Power Plants (NPPs) is a core factor in preventing human errors and ensuring the safe operation of NPPs. Due to the high uncertainty of human behaviors and the scarcity of relevant data, traditional analysis methods mainly rely on empirical judgment, which suffer from insufficient dynamics and poor engineering adaptability. To address the issues, this paper conducts a study on an AI prediction model for operator diagnosis level in a MCR of NPPs based on the SACADA database and machine learning technology. The model adopts a probabilistic neural network (PNN) as the main architecture, and proposes a hybrid method of network search considering density distribution combined with K-fold cross-validation, which breaks the traditional mode of a single smoothing factor adapting to an entire dataset. The analysis results show that the performance of the hybrid method proposed in this paper outperforms network search + K-fold cross-validation and particle swarm optimization + K-fold cross-validation methods in terms of accuracy, precision, recall, and F1-score. The five-fold cross-validation verifies that the model has good stability and good generalization ability. Further, the model is compared with common AI models such as BP neural network and RBF neural network. The results demonstrate that the proposed model has advantages in core indicators including overall accuracy (0.9444), macro-precision (0.9783), macro-recall (0.9063), and macro-F1-score (0.9362), and can effectively solve the problems of insufficient recognition of minority-class samples, overfitting, and underfitting. This research achieves professional and in-depth application of the SACADA database for diagnosis level prediction, extends existing research on prediction tasks, and delivers valuable theoretical insights and practical application significance. Full article

Real-time monitoring of radioactive liquid effluents from nuclear power plants (NPPs) is essential for environmental safety assurance, yet existing liquid scintillation counting (LSC) instruments are bulky (>200 kg), laboratory-bound, and incapable of autonomous online deployment. This paper presents the design and implementation of a fully automated online LSC monitoring process integrating seawater sampling, distillation pre-treatment, liquid scintillator mixing, dual photomultiplier tube (PMT) coincidence detection, field-programmable gate array (FPGA)-based digital signal processing, and 4G remote data transmission in a single portable unit weighing 21.59 kg. The automated process executes a complete sample-to-result cycle in approximately 45 min without human intervention. The signal processing chain comprises a dual-PMT coincidence system, a custom two-stage pre-amplifier, a 14-bit 40 MSPS analogue-to-digital converter (ADC; AD9245, Analog Devices, Norwood, MA, USA), and a five-stage FPGA pipeline implementing anti-coincidence rejection, pulse amplitude discrimination, charge comparison method (CCM) waveform discrimination, and convolutional neural network (CNN)-based alpha/beta classification achieving 97.4% accuracy on a Geant4-simulated test set. System performance was validated against a PerkinElmer 1220 QUANTULUS reference spectrometer across a five-point calibration range (0–400 Bq/L; R² = 0.9987, recovery 99.4–101.6%), confirmed via third-party environmental testing (−10 °C to +50 °C, GB/T 2423.1-2008), and verified in field measurements at Tianwan Nuclear Power Plant. The experimentally determined system background is (1.83 ± 0.12) cpm; the calculated minimum detectable activity (MDA) for tritium is 0.073 Bq/mL at 30 min counting time (η = 3.4%, V = 10 mL, T_s = 1800 s per the Currie formulation), satisfying the GB 14587 (the Chinese national standard: Limits of Radioactivity for Liquid Effluents from Nuclear Power Plant) regulatory reference limit of 0.5 Bq/mL with a 7× safety margin. The proposed system is, to the authors’ knowledge, the first reported instrument combining full process automation (including distillation pre-treatment), single-person portability, and real-time 4G remote data transmission for continuous NPP liquid effluent surveillance in high-salinity seawater environments. Full article

This study proposes a Time-Interval-Driven Sequence Construction (TIDSC) framework combined with a Transformer-based Autoencoder (Transformer-AE) for anomaly detection in nuclear power plant (NPP) Distributed Control System (DCS) controller logs. The proposed framework enhances log sequence construction by segmenting log streams according to temporal intervals, thereby helping preserve temporally coherent behavioral sequences. Based on the constructed behavior-oriented sequences, the Transformer-AE with a positional-only query decoder is trained using normal operational sequences through reconstruction-based learning and identifies anomalies based on reconstruction errors. Experimental results demonstrate improved performance over fixed-window and sliding-window baselines under the considered experimental settings. These results further suggest that temporal structure-aware sequence construction can improve behavioral representation learning and enhance anomaly separability in reconstruction error space. Overall, the proposed framework provides a potential approach for anomaly detection in the investigated NPP DCS controller log dataset, highlighting the value of integrating temporal sequence modeling with reconstruction-based learning. Full article

The planned Lubiatowo–Kopalino nuclear power plant (NPP) on the Polish Baltic coast requires a cooling technology that balances energy security, economic efficiency, and compliance with a multi-layered framework governing thermal discharge in a sensitive sea. This article integrates regulatory analysis across four instruments (EU Water Framework Directive, Polish discharge standards, HELCOM Baltic guidelines, and IAEA practice) with site-specific evidence from a 2D advection-diffusion thermal plume model. Four cooling options (once-through seawater, closed-loop towers, dry-air, and hybrid) are evaluated against regulatory criteria and against the documented vulnerabilities of the southern Baltic: eutrophication, restricted flushing, and ongoing warming. The modelling indicates that the acute thermal plume (ΔT ≥ 2 °C) remains limited to approximately 1.18 km² even under summer 90th-percentile conditions, whereas low-magnitude warming (ΔT = 0.1–0.5 °C) extends over approximately 1886.6 km². Although the once-through system with a multi-port diffuser satisfies current regulatory criteria, the extensive far-field anomaly represents an additional ecological stressor for an already eutrophic and warming marine ecosystem. A staged hybrid approach, with the first unit on validated once-through cooling and subsequent units on hybrid dry/wet systems, emerges as the most sustainable pathway, balancing empirical learning, regulatory foresight, and long-term climate resilience. The analysis offers a transferable framework for thermal-discharge assessment in shallow, microtidal, nutrient-enriched coastal seas, while demonstrating consistency with established regulatory practice and published knowledge on thermal-plume behaviour. Full article

Conventional analysis of nuclear power plant accident transient responses heavily relies on physical simulation programs, whose computational time significantly exceeds the actual accident response duration, thereby severely hindering real-time safety assessment. This paper proposes a novel framework that integrates a Transformer-based Variational Autoencoder (VAE) with knowledge distillation for the fast prediction of nuclear power plant (NPP) accident transient responses. The approach involves constructing a latent space to extract essential features from transient response data using an Encoder–Decoder model based on the VAE architecture. A key innovation is the establishment of a direct mapping between plant operating condition parameters and the latent space, enabling the one-step generation of accident transients without iterative sequential prediction. The Encoder and Decoder leverage Self-Attention and Cross-Attention mechanisms to enhance feature extraction and conditional generation. Furthermore, the Encoder is distilled into a Mapper network, which predicts the latent features directly from the operating conditions, resulting in an efficient Mapper–Decoder pipeline for rapid prediction. The proposed model was evaluated against traditional Long Short-Term Memory (LSTM) and fully connected neural networks. Experimental results demonstrate that the proposed framework achieves superior performance in predicting NPP accident transients, indicating its strong potential for efficient safety analysis and system design optimization. Full article

In safety-critical industrial environments such as nuclear power plants (NPPs), early detection of pipeline leakage is essential for maintaining operational safety. However, leakage events are rare, abnormal samples are difficult to collect, and obtaining sufficient condition-specific normal data is also challenging. To address these limitations, this paper proposes SiameseGAD, a generative few-shot anomaly detection framework for pipeline leakage detection in the secondary systems of NPPs. The proposed method formulates leakage detection as a few-shot normality-modeling problem rather than as a problem of directly learning anomaly patterns. The Siamese network learns similarity relationships among normal samples and constructs a normal feature manifold, while anomaly scores are computed based on the distance from the estimated normal distribution. To improve normal distribution estimation under limited data, a denoising diffusion probabilistic model (DDPM) is used to generate in-distribution normal variants to augment the support-set. The main contribution of SiameseGAD lies in combining metric-learning-based few-shot normality modeling with normal-to-normal generative augmentation, enabling anomaly detection using only a few normal samples without relying on real anomaly data or synthetic anomaly generation. In the three evaluated target classes, SiameseGAD achieved an average AUROC of 93.59% and an average accuracy of 95.26%. These results indicate the potential of SiameseGAD for few-shot anomaly detection using only normal support samples, without requiring real or synthetically generated anomaly samples during inference. Full article

Accident diagnosis in nuclear power plants (NPPs) should provide mechanism-aligned evidence that can be reviewed by operators and safety engineers, rather than only a high-confidence accident label. Existing data-driven methods achieve strong classification performance but often express explanations as attention maps, anomalous nodes, prototypes, or causal links separately, making it difficult to obtain a unified diagnostic evidence chain. To address this limitation, we propose a Concept-Constrained Physical Graph (CCPG) framework that formulates accident diagnosis as structured evidence generation. CCPG groups multivariate transient signals into operator-readable physical nodes, extracts node-wise temporal features, and propagates them over a mechanism-guided graph. It then couples a concept bottleneck with implicit latent features, prototype learning, and edge/stage supervision to predict the accident class and a reviewable evidence package. On the evaluated NPPAD five-class simulated benchmark, CCPG achieved saturated clean-set classification (1.000 accuracy) and high paired-challenge accuracy (0.996) while providing concept, affected-node, edge-template, stage-order, and prototype evidence. Additional analyses, including Transformer baselines, feature-restricted and early-window stress protocols, calibration, statistical testing, open-set detection, and scalability profiling, further characterize the robustness, reliability, and applicability of the proposed framework. Full article

Failures in critical systems and equipment within nuclear power plants (NPPs) significantly threaten operational safety and reliability. Therefore, rapid and accurate root cause localization during the incipient stages of failure is critical to preventing escalation. Traditional modeling methods often fail to address the inherent structural complexity of NPPs, the diversity of failure modes, and the stochastic mapping relationships between symptoms and causes. To address these challenges, this paper proposes an intelligent fault diagnosis framework integrating knowledge graphs (KGs) and Bayesian networks (BNs). First, by analyzing failure modes and anomaly characteristics, we define discrimination criteria for typical faults. Second, a structured knowledge modeling approach is developed to transform unstructured fault information into a KG, which is subsequently mapped to a BN topology. Finally, to mitigate the subjectivity of expert priors, data-driven structure and parameter learning algorithms are employed to optimize the model, enhancing inference accuracy. Robustness was validated through experiments targeting three fault severity levels, using signed directed graphs (SDGs), support vector machines (SVMs), domain generalization softmax (DG-softmax) and long short-term memory (LSTM) as benchmarks. Experimental results demonstrate that the proposed method maintains high diagnostic precision across varying severities, outperforming traditional data-driven methods in accuracy and stability. This study enhances the interpretability and engineering applicability of intelligent diagnosis in nuclear power systems. Full article

Alpha/beta (α/β) pulse shape discrimination (PSD) in liquid scintillation counting (LSC) is fundamentally limited by the charge comparison method (CCM) at low energies, where the entire tritium (³H) beta spectrum resides (0–18.6 keVee). The CCM figure-of-merit drops below 0.6 in this region, rendering it inadequate for simultaneous tritium and natural uranium alpha monitoring in nuclear power plant (NPP) liquid effluents. We present a one-dimensional convolutional neural network (1D-CNN) trained on an 80,000-waveform physics-based simulation dataset using established scintillation parameters for Ultima Gold AB. The proposed network achieves 97.4% overall classification accuracy and an area under the receiver operating characteristic curve (AUC) of 0.9981 on the held-out test set, representing improvements of 13.8 percentage points and 0.046 AUC over CCM. In the critical 0–18.6 keVee region, CNN accuracy exceeds 95% compared to below 60% for CCM—a greater than 35 percentage point improvement. Pulse amplitude discrimination (PAD), evaluated as a preliminary screening method, exhibits a 6.3% alpha spillover rate into the beta window, exceeding the regulatory limit of 3%. Gradient-weighted class activation maps (Grad-CAM) confirm that the network exploits physically meaningful pulse features rather than simulation artefacts. A comprehensive background suppression strategy combining dual-SiPM coincidence (24× reduction), anti-coincidence guard detector (5.8× reduction), composite passive shielding (10× reduction), and CNN-assisted discrimination reduces the system equivalent background to 1.83 ± 0.12 cpm, yielding a tritium minimum detectable activity (MDA) of 0.21 Bq/mL (10 mL sample, 30 min count), which satisfies the GB 14587 reference limit of 0.5 Bq/mL. After 8-bit post-training quantisation, the model achieves sub-microsecond inference latency on an embedded Xilinx Artix-7 Field-programmable gate array(FPGA), enabling real-time deployment in portable online monitoring systems. Full article

Cooling water systems of coastal nuclear power plants in China are frequently threatened by blockages caused by marine organisms. However, long-term studies on macrobenthic community dynamics and their associations with environmental factors are scarce, limiting the precise prevention of such blockage risks. This study conducted quantitative monitoring of macrobenthos and synchronous measurement of water environmental factors at 24 sampling stations in three functional areas (water intake, harbor basin, and drainage outlet) adjacent to the Northeast Fujian NPP from 2018 to 2024. Community structure characteristics were analyzed using the Shannon–Wiener and Margalef indices. The Grappler Method Risk Index (GMRI) was employed to screen species at risk of blocking cooling water systems, and the Mantel test and random forest models were applied to explore the associations between the macrobenthic community and environmental factors. A total of 161 macrobenthic species were identified. Polychaetes (71 species, accounting for 44.1%) were the absolute dominant group, followed by crustaceans (35 species) and Mollusks (30 species). The interannual fluctuation range of the polychaete proportion was 41.1–57.8%, reaching a peak in 2023. There were significant differences in community structure among different areas (PERMANOVA, p < 0.05), with the largest inter-regional difference in 2024 (R² = 0.36). The annual average number of species (9 species), density (155.25 ind./m²), and biomass (29.58 g/m²) in the drainage outlet were higher than those in the water intake and harbor basin. The GMRI identified Protankyra bidentata (spiny sea cucumber, GMRI values of 50.67% to 64.98% from 2019 to 2023) and Actiniaria sp. (sea anemone, a GMRI value of 54.63% in 2021) as medium-risk species for cooling water system blockage, while most other organisms were classified as low risk or extremely low risk. The Mantel test and random forest analysis confirmed that nitrogen nutrients (NO₃⁻) and phosphorus (PO₄³⁻) were significantly positively correlated with the polychaete community. Furthermore, NO₃⁻ and NH₄⁺ each explained 13.66% of the variation in the diversity index (H′), serving as key factors driving community structure. This study demonstrates the co-dominance of thermal and nutrient drivers in shaping macrobenthic communities over a multi-year scale, and identifies specific, morphologically suited taxa as potential blockage risks. The findings provide a scientific basis for targeted risk-species monitoring and support the integration of long-term ecological data into NPP cooling water system security management. Full article

Severe transients in nuclear power plants (NPPs) are strongly coupled and highly nonstationary, which makes reliable short-horizon multivariate forecasting difficult for conventional sequence models. To address this challenge, this study develops a hybrid LSTM-Transformer forecasting framework for severe nuclear accident time series and uses the Sparrow Search Algorithm (SSA) as a task-oriented joint hyperparameter optimization tool for nuclear accident forecasting. In this framework, the self-attention mechanism captures long-range temporal dependencies and cross-variable interactions, while the LSTM component strengthens the modeling of short-term dynamics and local temporal memory. SSA is employed as a task-oriented joint hyperparameter optimization tool to adapt key model settings, including the number of attention heads, encoder depth, model dimension, LSTM hidden units, and dropout rate, for severe nuclear accident forecasting. In addition, a regularized training strategy combining dropout and validation-based early stopping is adopted to alleviate overfitting and improve training stability. The main comparison results are reported as mean ± standard deviation over 20 independent runs with the same data split and different random seeds. Experiments on high-fidelity PCTran/APR1400 simulations covering LOCA, LACP, and SLBIC scenarios, together with a severity-shifted LOCA test, demonstrate strong and statistically stable predictive performance. Across the three representative accident scenarios, the proposed framework achieves mean ${\mathrm{R}}^2$ values of 0.943 ± 0.009, 0.951 ± 0.007, and 0.946 ± 0.010, while maintaining about 30% lower mean nRMSE and nMAE than the strongest LSTM-Transformer baseline. A 2 × 2 ablation study shows that regularization mainly improves training efficiency, reducing the required epochs by a range of about 36–41%, whereas SSA primarily improves predictive accuracy through better hyperparameter selection. Their combination provides the best overall generalization. Cross-severity LOCA evaluation further confirms the robustness of the proposed model, yielding mean ${\mathrm{R}}^2$ = 0.885 ± 0.017 and mean nRMSE = 0.100 ± 0.010. The model also achieves low inference latency (P50 = 7.6 ms per sample), indicating its computational potential for near-real-time multivariate forecasting in safety-critical transient monitoring. Full article

During the transition of global energy systems toward low-carbon and high-reliability operation, nuclear power plant (NPP) operation and maintenance require environmental perception methods that are safe, energy-efficient, and sufficiently accurate for confined and radiation-risk areas. To address these requirements, this paper proposes an energy-oriented multi-robot collaborative exploration and mapping framework, termed I-WFD-Gmapping-DT. The framework integrates a digital twin (DT) 5+3 model, improved wavefront frontier detection (I-WFD), energy- and risk-aware task allocation, EKF-AMCL-based initial relative pose estimation, and multi-scale Gmapping map fusion. Unlike conventional frontier-based or single-objective exploration methods, the proposed utility function jointly considers discounted information gain, obstacle-sensitive path cost, estimated battery energy, angular dispersion, and safety constraints. A ROS-Gazebo simulation of an NPP-like environment was used for 30 independent runs with randomized seeds and starting perturbations. Compared with WFD-Gmapping, the proposed method increased the three-robot coverage area percentage from 35.6 ± 2.1% to 40.5 ± 1.9%, reduced exploration time by 13.35%, reduced total and used frontier target points by 38.9% and 23.24%, respectively, and reduced estimated energy consumption by 13.9%. Map accuracy was also improved, with AE decreasing from 12.45% to 11.52%, RMSE from 7.85% to 7.18%, and SSIM increasing from 0.78 to 0.83. Additional sensitivity, ablation, runtime, and initial-pose experiments confirm the robustness of the parameter selection and the contribution of the DT-enabled feedback mechanism. The results show that I-WFD-Gmapping-DT can enhance collaborative inspection efficiency, reduce redundant motion and energy consumption, and provide reliable mapping support for intelligent NPP operation and maintenance. Full article

As a member state of the European Union, Bulgaria is committed to decarbonisation and the achievement of sustainable development goals. The country has a well-established energy sector and is a net exporter of electricity produced from diverse sources. Electricity generation relies mainly on two key pillars: lignite-fired Thermal Power Plants (TPPs) and the Nuclear Power Plant (NPP) in Kozloduy. This study examines the status of Bulgarian state-owned energy companies (SOEC) and their capacity to respond to the challenges of a sustainable transition towards low- or zero-emission electricity production. The study contributes to the existing literature by providing insights from a comparative analysis of state-owned thermal and nuclear power generation in Bulgaria, examined through the lens of sustainable development. From a practical standpoint it contributes by outlining possible pathways for the sustainable transformation of carbon-intensive TPPs. The analy-sis is based on key sustainability indicators covering the three pillars of sustainable development—economic, social and environmental performance. It includes not only an assessment of the financial performance of state-owned thermal power plants and the nuclear power plant over the past five years but also selected social and environmental indicators. The findings suggest that nuclear energy production in Bulgaria is largely consistent with the core principles of sustainability, while coal-based thermal power plants face increasing economic pressures and contribute to significant environmental impacts. The results highlight the need to transform the coal-based electricity sector into a more economically viable and socially responsible alternative, such as low-carbon generation technologies including nuclear energy. Full article

In recent years, frequent blockage of water intake structures at nuclear power plants (NPPs) by marine organisms has increased the risk of cooling source loss for the plants. Optimizing the layout of water intake structure to actively avoid or divert blockages near the intake entrance is one of the effective measures for cooling source risk prevention and control, and relevant research remains scarce at present. Taking a certain NPP as the research object, this paper simulates the flow field and particle transport in the sea area around the water intake based on a hydrodynamic-particle coupling model. A method for determining the maximum water source range and critical tidal conditions under risk source uncertainty is proposed. The flow pattern and entrainment risks of different open channel inlet types are compared. The results show that when the water intake open channel is arranged perpendicular to the ambient flow, a large recirculation zone exists at the intake entrance. Simply increasing the width at the intake entrance by expanding the local opening has an insignificant effect on reducing the water intake velocity and entrainment risk, while adopting additional side opening intake plays a certain role in dispersing the water intake entrainment intensity. The research results provide a basis for the optimal design and operation of water intake at NPPs. Full article