检索结果-内蒙古大学图书馆

Residual resampling-based physics-informed neural network for neutron diffusion equations

学校读者我要写书评

暂无评论

arXiv

arXiv 2024年

作者： Zhang, Heng He, Yun-Ling Liu, Dong Hang, Qin Yao, He-Min Xiang, Di Chongqing University of Posts and Telecommunications Chongqing400065 China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu610213 China CNNC Engineering Research Center of Nuclear Energy Software and Digital Reactor Chengdu610213 China

The neutron diffusion equation plays a pivotal role in the analysis of nuclear reactors. Nevertheless, employing the Physics-Informed Neural Network (PINN) method for its solution entails certain limitations. Traditional PINN approaches often utilize fully connected network (FCN) architecture, which is susceptible to overfitting, training instability, and gradient vanishing issues as the network depth increases. These challenges result in accuracy bottlenecks in the solution. In response to these issues, the Residual-based Resample Physics-Informed Neural Network(R2-PINN) is proposed, which proposes an improved PINN architecture that replaces the FCN with a Convolutional Neural Network with a shortcut(S-CNN), incorporating skip connections to facilitate gradient propagation between network layers. Additionally, the incorporation of the Residual Adaptive Resampling (RAR) mechanism dynamically increases sampling points, enhancing the spatial representation capabilities and overall predictive accuracy of the model. The experimental results illustrate that our approach significantly improves the model’s convergence capability, achieving high-precision predictions of physical fields. In comparison to traditional FCN-based PINN methods, R2-PINN effectively overcomes the limitations inherent in current methods, providing more accurate and robust solutions for neutron diffusion equations. Copyright © 2024, The Authors. All rights reserved.

关键词： Network architecture

Direct Numerical Simulation of Turbulent Flow and Heat Transfer in 3x3 Square Rod Bundle

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Miao, Heng Li, Jie Wei, Zonglan Sun, Yumeng Ai, Zhelun Deng, Jian Sino-French Institute of Nuclear Engineering and Technology Sun Yat-sen University Guangdong Zhuhai519082 China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Sichuan Chengdu610213 China

The flow characteristics within rod bundles are crucial in the evaluation of system performance and safety of engineering systems, including various nuclear reactor designs. This paper conducts direct numerical simulation (DNS) of a 3 × 3 square rod bundle using Nek5000. The model dimensions are Lx × Ly × Lz = 0.95 × 0.02045 × 0.02045 m, with a Reynolds number of 22876. The velocity field and temperature field are analyzed in detail, and the physical quantities such as velocity, temperature and Reynolds stress are extracted. The results show that the organized fluid motion is significant, indicating that there may be gap instability. Furthermore, an assessment of applicability for the widely used RANS model is provided for the rod bundle channel, including the recommended wall treatment method and turbulence model with strong applicability. © 2024, The Authors. All rights reserved.

关键词： Heat transfer

Research of the Neutronics Performance Improvement Based on the Small Lead-Based Fast reactor Slbr-50

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Zhao, Chen Lou, Lei Zhang, Bin Zhou, Bingyan Peng, Xingjie Wang, Lianjie Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu China

design of high-performance lead-based fast reactor has already become a hotspot in the advanced reactor system. Research of the neutronics performance improvement is presented based on the small lead-based fast reactor SLBR-50. Parameter sensitivity analyses are conducted, including height-diameter ratio(H/D), reflector assembly arrangement, pitch-diameter ratio(P/D) in fuel assembly, fuel material and fuel enrichment partition. Numerical results of eigenvalue in burnup procedure, assembly power distribution and energy spectrum are analyzed. These results provide influences of these design parameters in quantity and the neutronics performance can be improved with parameter optimization. © 2022, The Authors. All rights reserved.

关键词： Fast reactors

Deep learning-based dual monitoring system for power forecasting and fault detection in nuclear power applications

学校读者我要写书评

暂无评论

Energy and AI 2025年 20卷

作者： Lyu, Mingzhe Gong, Helin Chen, Zhang Wang, Jiangyu Zhong, Mingxiao Wang, Zhiyong Li, Qing Pan, Zefei School of Mathematical Sciences University of Electronic Science and Technology of China Chengdu611731 China Shanghai Zhangjiang Institute of Mathematics Shanghai201203 China Paris Elite Institute of Technology Shanghai Jiao Tong University Shanghai200240 China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu610041 China CNNP Operation Maintenance Technology Co. Ltd Hangzhou311215 China

Monitoring key parameters in nuclear power plant control rooms is critical, as human errors can result in severe safety and operational consequences. This study proposes a hybrid framework for power prediction and fault detection that integrates multi-head self-attention mechanisms with long short-term memory networks, combined with a dual-monitoring system. The framework is evaluated using real-time data from two pressurized water reactor units (Units 5 and 6) under four realistic operational scenarios. In the most informative case, the model achieves a 56.6% reduction in root mean square error and a 36.8% reduction in mean absolute error, with a coefficient of determination (R2) of 0.9924—significantly outperforming the next-best benchmark. For fault diagnosis, the dual-monitoring system reduces the false negative rate to 18.73% and improves recall to 81.27%, demonstrating strong anomaly detection under complex conditions. By combining short-term fluctuation sensitivity with long-term trend stability, the proposed approach offers a robust and generalizable solution for intelligent monitoring. These findings advance the development of artificial intelligence–enhanced systems for secure and efficient operation of critical energy infrastructure. © 2025 The Authors

关键词： Deep learning

POST-HOC INTERPRETABILITY BASED PARAMETER SELECTION for DATA ORIENTED NUCLEAR reactor ACCIDENT DIAGNOSIS SYSTEM

学校读者我要写书评

暂无评论

arXiv

arXiv 2022年

作者： Li, Chengyuan Li, Meifu Qiu, Zhifang Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu China

During applying data-oriented diagnosis systems to distinguishing the type of and evaluating the severity of nuclear power plant initial events, it is of vital importance to decide which parameters to be used as the system input. However, although several diagnosis systems have already achieved acceptable performance in diagnosis precision and speed, hardly have the researchers discussed the method of monitoring point choosing and its layout. For this reason, redundant measuring data are used to train the diagnostic model, leading to high uncertainty of the classification, extra training time consumption, and higher probability of overfitting while training. In this study, a method of choosing thermal hydraulics parameters of a nuclear power plant is proposed, using the theory of post-hoc interpretability theory in deep learning. At the start, a novel Time-sequential Residual Convolutional Neural Network (TRES-CNN) diagnosis model is introduced to identify the position and hydrodynamic diameter of breaks in LOCA, using 38 parameters manually chosen on HPR1000 empirically. Afterwards, post-hoc interpretability methods are applied to evaluate the attributions of diagnosis model's outputs, deciding which 15 parameters to be more decisive in diagnosing LOCA details. The results show that the TRESCNN based diagnostic model successfully predicts the position and size of breaks in LOCA via selected 15 parameters of HPR1000, with 25% of time consumption while training the model compared the process using total 38 parameters. In addition, the relative diagnostic accuracy error is within 1.5% compared with the model using parameters chosen empirically, which can be regarded as the same amount of diagnostic reliability. Copyright © 2022, The Authors. All rights reserved.

关键词： Nuclear power plants

Study on Diffusion and Migration Characteristics of Impurities in Secondary Side of Simplified U-Tube Steam Generator

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Li, Meng Cheng, Kun Wen, Jiming Tian, Ruifeng Fundamental Science on Nuclear Safety Simulation Technology Laboratory Harbin Engineering University Harbin150001 China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu610213 China

In the operation process of nuclear power system, leakage accidents of main and auxiliary condensers are easy to occur, which leads to seawater entering the secondary loop from the condenser, and impurities such as Cl- enter the steam generator with the feed water of the secondary loop, and are enriched near the heat transfer tube. It is easy to cause wear and corrosion of steam generator heat transfer tube, which threatens the safety of nuclear power plant. In order to study the migration and diffusion process of impurity Cl- in the secondary side of steam generator, a multi-bundle three-dimensional physical model of fluid-solid coupling heat transfer in the primary and secondary loop of steam generator and heat transfer tube was established respectively. The fluid flow and heat transfer process between Ushaped tube bundles were simulated numerically to reveal the distribution law of key thermo-hydraulic parameters between Ushaped tube bundles, such as volume vapor fraction, temperature, velocity, etc. The diffusion and migration of Cl- in two-phase flow were also investigated. The influence of thermalhydraulic parameters on impurity migration, two-phase flow in the secondary side and effective diffusion coefficient of chloride ions on chloride ion concentration distribution were *** concentration of chloride ions changes with the change of effective diffusion coefficient of chloride ions. The effective diffusion coefficient of chloride ions is affected by the boiling intensity of coolant and the flow velocity of coolant. The synergistic effect of the boiling intensity of coolant and the flow velocity of coolant affects the migration and diffusion of chloride *** above phenomena may lead to impurity deposition and stress corrosion of heat transfer tubes near the surface. © 2023, The Authors. All rights reserved.

关键词： Steam generators

Uncertainty Quantification of Spray Nozzle Fatigue Analysis Under Transient Load 1

学校读者我要写书评

暂无评论

作者： Liu, Zhenyu Li, Lijuan Zhang, Liping Li, Hui Shao, Xuejiao Jiang, Lu Zhang, Ying Science and Technology On Reactor System Design Technology Laboratory Nuclear Power Institute of China No. 328 Changshun Avenue Chengdu Sichuan610200 China

ISBN: (数字)9783030683030

ISBN: (纸本)9783030683023;9783030683054

In this paper, the randomness concept of input parameters is introduced into the design of nuclear equipment, and the uncertainty analysis of input parameters is carried out for fatigue analysis. Taking the pressurizer spray nozzle as the research object, considering the random fluctuation of temperature and pressure of the transient load, the reliability and distribution of the fatigue usage are obtained by performing fatigue analysis according to the fatigue analysis method of nuclear power equipment with Monte Carlo sampling method. Taking section No.3 as an example, the thermal analysis and stress analysis models are established. Through the parameter sensitivity analysis, the relative sensitivity of the fatigue usage to each input parameter is obtained. From the results, it is clear which parameters have a significant impact on the results. The results can further optimize the structural design of the nuclear power equipment and improve its fatigue resistance, which has great application value to the life extension of nuclear power equipment. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Uncertainty analysis

Experimental Study on Flow Resistance and Phase Distribution of Gas-Liquid Flow in Helical Cruciform Fuel Assembly

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Zhang, Qi Wang, Haoyu Li, Junlong Ren, Quanyao Zhou, Yi Zheng, LeLe Zhang, Yile Li, Dongyang Gu, Hanyang Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu610041 China School of Nuclear Science and Engineering Shanghai Jiao Tong University Shanghai200240 China Harbin Engineering University China

The helical cruciform fuel (HCF) is characterized by self-supporting and enhanced transverse mixing, which may replace the cylindrical fuel of the light water reactor (LWR). In this work, the flow resistance and phase distribution of a 4×4 HCF assembly was investigated by experimental method. The experimental system of air-water two-phase flow was established, and the void fraction was gained by wire-mesh sensor (WMS). The dummy HCF elements produced by additive manufacturing technology were assembled with the locating plates. The evolution of the flow regime and the void fraction of the central and peripheral were paid particular attention. The correlations for the two-phase friction multiplier and void fraction were proposed based on the experimental data. © 2024, The Authors. All rights reserved.

关键词： Two phase flow

Peridynamic Simulation for Thermal and Mechanical Behavior of Manufactured Structures with MIG Welding 1

学校读者我要写书评

暂无评论

作者： Li, Hui Zhang, Liping Liu, Zhenyu Li, Lijuan Zhang, Yixiong Shao, Xuejiao Science and Technology On Reactor System Design Technology Laboratory Nuclear Power Institute of China No. 328 Changshun Avenue Chengdu Sichuan610200 China

ISBN: (数字)9783030683030

ISBN: (纸本)9783030683023;9783030683054

A new numerical simulation method is developed for the evaluation of thermal and mechanical analyses of manufactured structures with Metal Inert Gas (MIG) welding on the basis of the peridynamics. The peridynamic formulations of the weak coupled thermo-mechanical problems are derived based on a heat transfer model and a microplastic model. To simulate the MIG welding process, a birth–death material point method in the peridynamic framework is proposed to model the addition of weld metal to the workpiece. To simplify the calculation process and improve the computational efficiency due to no need of updating the neighbourhoods of material points, an additional scalar field is introduced in the equations to describe the birth and death states of the material points. In addition, the present method is applied to simulate a classical model of two plates being joined together by the MIG welding. The results indicate that the residual tensile deformations are formed in the welded area while the residual compressive deformations are formed far from the wild line after the release of the clamps on account of the plastic deformation. These results also demonstrate the validity of the present method for the simulation of the MIG welding by comparing with the reference solution. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Computational efficiency