检索结果-南通市图书馆

SensFL:Privacy-Preserving Vertical Federated Learning with Sensitive Regularization

在线全文

学校读者我要写书评

暂无评论

computer Modeling in engineering & Sciences 2025年第1期142卷 385-404页

作者： Chongzhen Zhang Zhichen Liu Xiangrui Xu Fuqiang Hu Jiao Dai Baigen Cai Wei Wang School of Automation and Intelligence Beijing Jiaotong UniversityBeijing100044China Shuohuang Railway Development co. Ltd.National Energy GroupCangzhou062350China Beijing Key Laboratory of Security and Privacy in Intelligent Transportation School of Computer Science and TechnologyBeijing Jiaotong UniversityBeijing100044China

In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data *** Federated Learning(VFL)has emerged as a promising approach to facilitate such collaboration,allowing diverse entities to collectively enhance machine learning models without the need to share sensitive training ***,existing works have highlighted VFL’s susceptibility to privacy inference attacks,where an honest but curious server could potentially reconstruct a client’s raw data from embeddings uploaded by the *** vulnerability poses a significant threat to VFL-based intelligent railway transportation *** this paper,we introduce SensFL,a novel privacy-enhancing method to against privacy inference attacks in ***,SensFL integrates regularization of the sensitivity of embeddings to the original data into the model training process,effectively limiting the information contained in shared *** reducing the sensitivity of embeddings to the original data,SensFL can effectively resist reverse privacy attacks and prevent the reconstruction of the original data from the *** experiments were conducted on four distinct datasets and three different models to demonstrate the efficacy of *** results show that SensFL can effectively mitigate privacy inference attacks while maintaining the accuracy of the primary learning *** results underscore SensFL’s potential to advance privacy protection technologies within VFL-based intelligent railway systems,addressing critical security concerns in collaborative learning environments.

关键词： Vertical federated learning privacy defenses

A 2.5-dimensional Analytical Model of cold Leveling for Plates with Transverse Wave Defects

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Iron and Steel Research International 2015年第8期22卷 664-671页

作者： Wei-hua CHEN Juan LIU Zhen-shan CUI Ying-jie WANG Ying-rui WANG National Die & Mold CAD engineering Research Center Shanghai Jiaotong University baosteel co. Ltd.

Waves occurring in cold-rolled plates or sheets can be divided into longitudinal and transverse waves. Classical leveling theories merely solve the problem of longitudinal waves, while no well accepted method can be employed for transverse waves. In order to investigate the essential deformation law of leveling for plates with transverse waves, a 2.5-dimensional （2.5- D） analytical approach was proposed. In this model, the plate was transversely divided into some strips with equal width; the strips are considered to be in the state of plane strain and each group of adjacent strips are assumed to be deformation compatible under stress. After calculation, the bending deformation of each strip and the leveling effect of overall plate were obtained by comprehensNe consideration of various strips along with the width. Bending of roller is a main approach to eliminate the transverse waves, which is widely accepted by the industry, but the essential effect of bending of roller on the deformation of plates and the calculation of bending of roller are unknown. According to the 2.5-D analytical model, it can be found that, for plates, it is neutral plane offsetting and middle plane elongation or contraction under inner stress that can effectively improve plate shape. Taking double side waves as an example, the appropriate values of bending of roller were obtained by the 2.5-D analytical model related to different initial unevenness, which was applicable to the current on-line adjusting of bending of roller in rolling industry.

关键词： plate leveling 2.5-dimensional analytical model transverse wave middle plane deformation

Mechanism of separating pyrite and dolomite by flotation

在线全文

学校读者我要写书评

暂无评论

Journal of University of Science and Technology Beijing 2007年第4期14卷 291-296页

作者： Anping Liu Wen Ni Wei Wu Civil and Environmental engineering School University of Science and Technology Beijing Beijing 100083 China Nanjing Meishan Mining co. ltd. Baosteel Group Nanjing 210041 China Key Lab of the Ministry of Education of China for Nanomaterials Beijing University of Chemical Technology Beijing 100029 Chin

To study the mechanism of separating pyrite and dolomite by flotation, the acting mechanisms of WHL depressor and both the minerals were studied by means of thermogravimetric and differential thermal analysis （TG-DTA）, Fourier transform infrareddiffuse reflection spectroscopy （PTIR-DRS）, and X-ray photoelectron spectroscopy （XPS）. The results indicated that WHL formed metal salts with metal ions dissolved in water from dolomite and pyrite, which then deposited on their surfaces. Both of the minerals could be depressed by WHL. In the process of flotation, sulfur was created besides the WIlL being absorbed on the surface of the sulfur concentrate, and its recovery rate was slightly affected.

关键词： flotation dolomite pyrite depressor mechanism

Internal Friction on the Bake-Hardening Behavior of 0.11C-1.67Mn-1.19Si TRIP Steel

在线全文

学校读者我要写书评

暂无评论

Journal of Materials Science & Technology 2009年第1期25卷 141-144页

作者： Renyu Fu Yu Su Ping Ye Xicheng Wei Lin Li Jicheng Zhang School of Materials Science and engineering Shanghai University Shanghai 200072 China Shanghai University of engineering Science Shanghai 200336 China Automotive engineering Academy of SAIC Motor co. Ltd. Shanghai 201206 China Shanghai baosteel Plates co. Ltd. Shanghai 201900 China

The bake-hardening （BH） values and the ＂internal friction-temperature＂ spectrums were studied for the baked 0.11C-1.67Mn-1.19Si TRIP （transformation induced plasticity） specimens with 0%, 2% and 6% prestrain. Results show that the experimental TRIP steel deserves good bake-hardening ability and cottrell atmosphere is the reason for its bake hardening characteristic. It is also concluded that both the number and the saturation degree of cottrell atmosphere might affect the BH value of TRIP steel.

关键词： TRIP steel BH value Internal friction

Development and commercial application of baosteel sintering flue gas recirculating process

在线全文

学校读者我要写书评

暂无评论

baosteel Technical Research 2014年第4期8卷 1-8页

作者： CUI Jian LI Xianwei SHI Lei DU Hongjin JIA Xiufeng XI Saifeng YU Bo WANG Ruyi YU Yongmei JIN Chun YANG Yun baosteel group corporation Research Institute Baoshan Iron & Steel Co.Ltd. baosteel Stainless Steel co. Ltd. Ningbo Iron & Steel co. Ltd. baosteel engineering & Technology group co. Ltd.

Because of the low temperature, large waste gas volume, high pollutant content, and complicated compositions, waste heat recovery and emission gas treatment of sintering flue gas have always presented a challenge in the steel industry, and this issue has attracted widespread attention both locally and abroad. Recently, based on the first domestic pilot plant and demonstration project, baosteel has performed a series of innovative research and development investigations on its sintering flue gas recirculating （SFGR） process,in which the system design and optimization, dioxin-related contaminant source suppression, ore matching structure optimization, wear- resistant design of cycling fans and pipelines, high-efficiency dust removal equipment, system control and stable operation strategy, flue gas mixing and switching control, circular hood sealing, oxygen content conditioning, recirculation sintering system process control, and model development have been studied, and a complete set of equipment and technology for the SFGR process has been initially developed. The investigation results suggest that the SFGR process can not only significantly reduce exhaust gas volume and pollutant emissions but also recover low-temperature waste heat and reduce the energy consumption of sintering; therefore, the overall technologies possess great value in energy savings ,pollution emission reduction, and sintering ore quality/yield improvement.

关键词： SFGR process sintering waste gas recirculation pilot plant demonstration project engineering application

Explainable machine learning model for predicting molten steel temperature in the LF refining process

在线全文

学校读者我要写书评

暂无评论

International Journal of Minerals,Metallurgy and Materials 2024年第12期31卷 2657-2669页

作者： Zicheng Xin Jiangshan Zhang Kaixiang Peng Junguo Zhang Chunhui Zhang Jun Wu Bo Zhang Qing Liu State Key Laboratory of Advanced Metallurgy University of Science and Technology BeijingBeijing 100083China School of Automation and Electrical engineering University of Science and Technology BeijingBeijing 100083China Tangshan Branch Hebei Iron and Steel Co.Ltd.Tangshan 063000China Xinjiang Bayi Iron and Steel co. Ltd.Baosteel GroupUrumqi 830022China Lvliang Jianlong Industrial co. Ltd.Lvliang 032100China

Accurate prediction of molten steel temperature in the ladle furnace(LF)refining process has an important influence on the quality of molten steel and the control of steelmaking *** research on establishing models to predict molten steel temperature has been ***,most researchers focus solely on improving the accuracy of the model,neglecting its *** present study aims to develop a high-precision and explainable model with improved reliability and *** eXtreme gradient boosting(XGBoost)and light gradient boosting machine(LGBM)were utilized,along with bayesian optimization and grey wolf optimiz-ation(GWO),to establish the prediction *** performance evaluation metrics and graphical representations were applied to compare the optimal XGBoost and LGBM models obtained through varying hyperparameter optimization methods with the other *** findings indicated that the GWO-LGBM model outperformed other methods in predicting molten steel temperature,with a high pre-diction accuracy of 89.35%within the error range of±5°*** model’s learning/decision process was revealed,and the influence degree of different variables on the molten steel temperature was clarified using the tree structure visualization and SHapley Additive exPlana-tions(SHAP)***,the explainability of the optimal GWO-LGBM model was enhanced,providing reliable support for prediction results.

关键词： ladle furnace refining molten steel temperature eXtreme gradient boosting light gradient boosting machine grey wolf op-timization SHapley Additive exPlanation

Modeling of LF refining process:a review

在线全文

学校读者我要写书评

暂无评论

Journal of Iron and Steel Research International 2024年第2期31卷 289-317页

作者： Zi-cheng Xin Jiang-shan Zhang Kai-xiang Peng Jun-guo Zhang Chun-hui Zhang Qing Liu State Key Laboratory of Advanced Metallurgy University of Science and Technology BeijingBeijing100083China School of Automation and Electrical engineering University of Science and Technology BeijingBeijing100083China Tangshan Branch Hebei Iron and Steel Co.Ltd.Tangshan063000HebeiChina Xinjiang Bayi Iron and Steel co. Ltd.Baosteel GroupWulumuqi830022XinjiangChina

With the increasing demand for energy conservation and emission reduction,more attentions have been paid to the intelligentization,greenization and low carbonization during the transformation and upgrading of steelmaking *** furnace(LF)refining is one of the key procedures in steelmaking process and has been widely used in steelmaking plants for its high equipment matching degree,low equipment investment and outstanding refining *** to the main tasks of LF refining process,the modeling methods of temperature prediction model,slag-making model,alloying model,argon blowing model and model of inclusions behavior were systematically reviewed,and the advantages and disadvantages of each modeling method were *** addition,the technical framework for the future has also been proposed based on existing works,including classification of raw materials,graphic representation of knowledge,introduction,upgradation and management of device/equipment,customization of steelmaking,modeling of refining process,synergy of models,intelligentization of decision-making,automation of control,and digitization of processes and operations,aiming to provide a reference for the modeling and intelligent development of LF refining process.

关键词： LF refining Temperature prediction Slag-making Alloying Argon blowing Inclusion behavior Modeling method

Microstructure and Mechanical Property of the Large Cross-Sectioned Mg-Gd-Y-Zn-Zr Alloy Produced by Small Extrusion Ratio

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Acta Metallurgica Sinica(English Letters) 2024年第6期37卷 999-1006页

作者： Sen Wang Hucheng Pan Caixia Jiang Zhihao Zeng Zhen Pan Weineng Tang Chubin Yang Yuping Ren Gaowu Qin Key Laboratory for Anisotropy and Texture of Materials(Ministry of Education) College of Materials Science and EngineeringNortheastern UniversityShenyang110819China Mg and Mg Alloy Institute Technology CenterBaosteel Metal Co.LtdChina Baowu Steel GroupShanghai200940China Jiangxi Provincial engineering Research Center for Magnesium Alloy Gannan Normal UniversityGanzhou341000China

In this work,the microstructure and mechanical properties of large cross-sectioned Mg-9Gd-3Y-1.2Zn-0.5Zr(VWZ931)samples produced by the small extrusion ratio has been *** as-extruded VWZ931 sample with diameter of~30 mm can exhibit the high yield strength(YS)of 339 MPa,ultimate tensile strength(UTS)of 387 MPa and elongation of 8.2%,*** peak-aged,the YS and UTS of the Mg samples were significantly increased to 435 MPa and 467 *** small extrusion ratio leads to the low fraction of dynamic recrystallized(DRX)grains in VWZ931 sample,and the texture hardening effect can be fully utilized to achieve high *** combined effect of precipitation strengthening due to the long-period stacking ordered phases and theβ′phase,grain boundary strengthening due to the fine DRX grains,heterogeneous deformation-induced strengthening caused by bimodal microstructure,can together contribute to the high strength of present Mg *** findings can shed light on designing other large-sized Mg wrought alloys with high mechanical performance.

关键词： Mg-RE alloy Mechanical property Bimodal grains Texture hardening

Influence of Monoclinic Zirconia Powder on the Properties of Zirconia Metering Nozzle Materials

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Influence of Monoclinic Zirconia Powder on the Properties of...