检索结果-内蒙古大学图书馆

Deep Learning Based Fault Diagnosis for Chemical Process with Statistical Feature Fusion

学校读者我要写书评

暂无评论

Deep Learning Based Fault Diagnosis for Chemical Process wit...

Chinese Control Conference (CCC)

作者： Lei Zhang Hao Xia Bao Cha School of Control Science and Control Engineering Dalian University of Technology Dalian P. R. China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu P. R. China

The existing fault diagnosis methods based on deep neural networks mostly rely on process data for fault classification and diagnosis, which makes it difficult to distinguish minor faults with similar data features. Therefore, this paper proposes a deep learning-based fault diagnosis model that combines statistical indicators and neural networks. By analyzing and calculating the relevant information entropy to screen the statistical signal features of time domain and frequency domain, we can extract more hidden information from the raw data, and use the convolutional autoencoder to compress the statistical indicators for feature extraction, and use CNN network to extract features from raw data. Finally, the two sets of features are fused and input into the LSTM network to obtain temporal information for fault diagnosis. Tests conducted on the Tennessee Eastman (TE) process show that the proposed model outperforms typical classifiers and improves the diagnostic accuracy of minor faults.

关键词：

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

Corrosion Mechanisms and Differences of Inconel 617 and Incoloy 800h Under High-Temperature Air Ingress Accident

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Li, Haoxiang Yan, Dapeng Li, Guodong Tian, Changqing An, Jiayi Tan, Jiaqi Zheng, Wei Du, Bin Yin, Huaqiang Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu610213 China State Key Laboratory of Advanced Nuclear Energy Technology Nuclear Power Institute of China Chengdu China Institute of Nuclear and New Energy Technology Key Laboratory of Advanced Reactor Engineering and Safety of Ministry of Education Tsinghua University Beijing China Tsinghua University China

In the high-temperature operation of very high-temperature gas-cooled reactors (VHTRs), "microclimate reaction" are one of the key corrosion reactions of concern. This study investigates the corrosion behavior and mechanisms of the two mainstream high-temperature alloys, Incoloy 800H and Inconel 617, under minor air ingress accidents. The main focus is on comparing and analyzing the "microclimate reaction" corrosion behavior and mechanisms of the two alloys, as well as exploring changes in their mechanical properties. The results show that during the experiment, both alloys primarily underwent oxidation and "microclimate reaction". The double-layer oxide of Cr2 O3 -MnCr2 O4 /MnFe2 O4 and the SiO2 layer on the surface of Incoloy 800H provide good resistance to "microclimate reaction," whereas the single-layer Cr2 O3 oxide on Inconel 617 is not effective in preventing "microclimate reaction". The mechanical property test results indicate that the decarburization caused by "microclimate reaction" does not significantly affect the basic mechanical properties of the alloy. Research on "microclimate reactions" should primarily focus on preventing the oxidation layer from cracking and reducing the time during which the alloy matrix is unprotected by a protective oxide layer. © 2024, The Authors. All rights reserved.

关键词： Gas cooled reactors

Dynamic Rod Worth Estimation with Extended Kalman Filter and Background Current Elimination

学校读者我要写书评

暂无评论

Dynamic Rod Worth Estimation with Extended Kalman Filter and...

第40届中国控制会议

作者： Hao Xia Junqi Yang Chao Bao Dalian University of Technology Science and Technology on Reactor System Design Technology Laboratory

The dynamical rod worth method(DRWM) is a fast and accurate method to measure and validate the worth of control rod during the zero power physics test at pressure water nuclear *** control rod worth is obtained from the reactivity change caused by the fast insertion of the control rod into the bottom of the *** reactivity is calculated based the measured ex-core detector signals which normally varies from～10 nA to～10 nA during the rod insertion *** the later stage of the test,as the detector signal becomes～10 nA,the impact of the measurement noise and the gamma-ray background current on the reactivity calculation becomes *** this paper,we developed an extended Kalman filter(EKF) with suitable weighing matrices to filter out the measurement *** most measurement noise filtered out,a background current estimation method is used to eliminate the effect of the background *** combining the background current estimation method with EKF filter,accurate control rod worth is obtained from the real test data.

关键词： Dynamic Rod Worth Method Reactivity Extended Kalman filter Background Current

Understanding Xenon and Vacancy Behaviors in Uo2, Un and U3si2: A Comparative Ab Initio Study

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Zhao, Jiajun Sun, Dan Wang, Yuanyuan Xi, Liu Chen, Ping Zhao, Jijun Dalian University of Technology Liaoning Dalian China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Sichuan Chengdu China

Extensive attention has been paid to accident tolerant fuels (ATFs), such as uranium mononitride (UN) and uranium sesquisilicide (U3Si2), which are regarded as the potential candidates to replace uranium dioxide (UO2) in light water reactor (LWR). However, the thermodynamic behavior of fission gas atom in these fuels that can quantitatively affect the burnup characteristics of ATF needs to be explored. To this end, the systematic ab initio investigation on the energetic properties of xenon (Xe)-vacancy complexes in UO2, UN and U3Si2 is performed with the GGA+U approach as well as the corrected chemical potential. The stabilities of Xe-vacancy clusters, including interstitial trap site (IS), mono-, bi- and tri-atomic vacancy, are thoroughly assessed. The formation energies of vacancy complexes indicate that it is more likely to form vacancy cluster defects and their complexes with Xe in UO2 and to generate mono-atomic vacancy and Xe-vacancy complex in both UN and U3Si2. Xe can be strictly confined by the trap sites in UO2 and UN, and yet in U3Si2, it prefers to move to the centre of a large free volume trap site. The strongest soluble characteristic of Xe in U3Si2 indicates excellent storage compacity of fission gas products in the matrix. This work can provide comprehensive insights into the original reason of interplay between Xe and vacancy as well as the thermodynamic behavior of those defects in the uranium-based fuels. © 2022, The Authors. All rights reserved.

关键词： Uranium dioxide

Direct Numerical Simulation of Turbulent Flow and Heat Transfer in a Rectangular Duct with High Aspect Ratio

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Sun, Yumeng Li, Jie Wei, Zonglan Ai, Zhelun Miao, Heng 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

Plate-type fuel is widely used due to its strong heat transfer capability, and its narrow rectangular geometry has different effects on flow heat transfer than traditional rectangular channels. Therefore, in this work, In this work, Direct Numerical Simulation(DNS) of a narrow rectangular duct was performed with the aim of analyzing flow and heat transfer phenomena and constructing a database of turbulent flow. The open-source program Nek5000 based on the spectral element method is used, the geometry is 0.57×0.00367×0.05 m with a high Aspect Ratio(AR) of 13.6, and the bulk Reynolds number is 24000 with a heat flux of 560 kW/m2. Common turbulence statistics like the time mean field, reynolds stress, turbulent heat flux distribution profile, and turbulent kinetic energy budget are obtained. The distribution of Reynolds stress and turbulent kinetic energy budget was analyzed, and the impact of high aspect ratio and material properties on flow and heat transfer was investigated. Additionally, the effects of single-sided and double-sided heating were studied. The turbulent flow database obtained from this study will be utilized for the later evaluation of turbulence models. © 2023, The Authors. All rights reserved.

关键词： Direct numerical simulation

Effect of rotationally accelerated shot peening on the microstructure and mechanical behavior of a metastableβtitanium alloy

学校读者我要写书评

暂无评论

Journal of Materials science & technology 2021年第16期75卷 27-38页

作者： Xinkai Ma Zhuo Chen Dongling Zhong SNLuo Lei Xiao Wenjie Lu Shanglin Zhang Key Laboratory of Advanced Technologies of Materials(Ministry of Education) School of Materials Science and EngineeringSouthwest Jiaotong UniversityChengdu 610031China State Key Laboratory of Solidification Processing School of Materials Science and EngineeringNorthwestern Polytechnical UniversityXi'an 710072China National Key Laboratory of Reactor System Design Technology Nuclear Power Institute of ChinaChengdu 610041China

Microstructural evolution and deformation mechanism of a metastableβalloy(Ti-10 V-2 Fe-3 Al)processed by rotationally accelerated shot peening(RASP)were systematically investigated with optical microscopy,X-ray diffraction,electron backscatter diffraction and transmission electron *** gradient hierarchical microstructures(gradients inα″martensite andβphase,and hierarchical twins range from the nanoscale to microscale)can be fabricated by RASP via changing the shot peening *** hardening behavior and tensile mechanical properties of gradient hierarchical microstructure were systematically *** deformation twinning systems of{112}α″and{130}<310>α″in the kinkedα″martensite were revealed during the tensile *** was found that stress-induced martensitic transformation,twinnedα″martensite and the related dynamic grain refinement contribute to hardness and work hardening *** improvement of strength and ductility of the metastableα″titanium alloy can be achieved by introducing a gradient hierarchical microstructure.

关键词： Titanium alloys Gradient hierarchical microstructure α″martensite Rotationally accelerated shot peening Mechanical behavior

Corrosive Effect on Saturated Pool Boiling Heat Transfer Characteristics of Metallic Surfaces with Hierarchical Micro/Nano Structures

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Xu, Wei Tang, Longchang Zhao, Ningkang Ouyang, Kun Zeng, Chang Liu, Xiaojing He, Xiaoqiang College of Smart Energy Shanghai Jiao Tong University Shanghai200240 China Shanghai200240 China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu610041 China

Surface modification is of critical interest to enhance boiling heat transfer in terms of heat transfer coefficient or critical heat flux (CHF), which is significantly affected by distinct surface morphology and wettability and it can improve the efficiency and safety of equipment. Furthermore, actual service environment may cause severe corrosion to the processed structured surfaces while its consequence on boiling heat transfer is still obscure. In this research, comprehensive researches are conducted to unravel corrosive effect on metallic samples made of stainless steel (SS) and Inconel materials with microstructures. Different constructions (i.e., microgroove, microcavity and micropillar array) and characteristic dimensions (~20, 50 μm) of microstructure, various duration time (up to 300 days) and pH values (~7.0 to 8.5) of corrosive environment are compared thoroughly. Conclusions can be drawn that not all microstructures can enhance pool boiling heat transfer characteristics, especially in terms of CHF values. More importantly, CHF value of SS microgroove sample firstly increases from 60.94 to 94.09 W·cm-2 in 50 days, then decreases to 47.77 W·cm-2 in 300 days, which can be attributed to competition result between formation of hierarchical micro/nano structure with enhancing wicking capability and chemistry condition with increasing contact angle. In addition, distinct bubble dynamics during pool boiling is also analyzed. The insights obtained from this article can be used to guide surface modification method and to reveal evolvement rule of engineered metallic surface in highly corrosive and harsh boiling heating transfer environment. © 2023, The Authors. All rights reserved.

关键词： Corrosive effects

Larger Recovery Strain in a Conisi Alloy Due to Enhanced Reversibility of FccHcp Martensitic Transformation

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Liao, Qi Wang, Dian Yang, Xiao He, Wei Yang, X. Peng, H.B. Wen, Y.H. School of Mechanical Engineering Sichuan University Chengdu China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu China School of of Mechanical Engineering Hebei University of Technology Tianjin China School of Manufacturing Science & Engineering China

It is not clear how coherent ordered nanoparticles affect characteristics of FCCHCP martensitic transformation in disordered alloys. In this paper, the precipitations of coherent ordered nanoparticles and their effects on the FCCHCP martensitic transformation were systematically investigated in a solution-treated Co24Ni6Si shape memory alloy. The results showed that both the coherent ordered L12 structured and L11 structured nanoparticles could precipitate even in the solution-treated Co24Ni6Si alloy followed by water quenching. The L12 structured nanoparticles strongly hindered the occurrence of stress-induced HCP martensitic transformation and its reverse transformation, but their hindrance dramatically reduced with the decrease of size when below 1.5 nm. The coherent ordered nanoparticles did not convert the nonthermoelastic FCCHCP martensitic transformation to the thermoelastic ones owing to absence of group-subgroup relation between the parent phase and martensite. A maximal recovery strain of 6.7% was obtained after only aging at 1073 K for 1 min. This value was 1~2% higher than the reported maximal ones in conventional FeMnSi-based alloys subjected to complicated thermomechanical treatments. The enhanced reversibility of HCP martensitic transformation could account for this larger recovery strain by the ultrafine coherent ordered L12 structured nanoparticles. © 2023, The Authors. All rights reserved.

关键词： Martensitic transformations