检索结果-南通市图书馆

Reinforcement Learning Based Quantization Strategy Optimal Assignment Algorithm for Mixed Precision

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, materials & Continua 2024年第4期79卷 819-836页

作者： Yuejiao Wang Zhong ma Chaojie Yang Yu Yang lu Wei Xi’an Microelectronics Technology Institute Xi’an710065China

The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to databit width. Reducing the data bit width will result in a loss of accuracy. Therefore, it is difficult to determinethe optimal bit width for different parts of the network with guaranteed accuracy. Mixed precision quantizationcan effectively reduce the amount of computation while keeping the model accuracy basically unchanged. In thispaper, a hardware-aware mixed precision quantization strategy optimal assignment algorithm adapted to low bitwidth is proposed, and reinforcement learning is used to automatically predict the mixed precision that meets theconstraints of hardware resources. In the state-space design, the standard deviation of weights is used to measurethe distribution difference of data, the execution speed feedback of simulated neural network accelerator inferenceis used as the environment to limit the action space of the agent, and the accuracy of the quantization model afterretraining is used as the reward function to guide the agent to carry out deep reinforcement learning training. Theexperimental results show that the proposed method obtains a suitable model layer-by-layer quantization strategyunder the condition that the computational resources are satisfied, and themodel accuracy is effectively *** proposed method has strong intelligence and certain universality and has strong application potential in thefield of mixed precision quantization and embedded neural network model deployment.

关键词： Mixed precision quantization quantization strategy optimal assignment reinforcement learning neural network model deployment

A novel algorithm for evaluating cement azimuthal density based on perturbation theory in horizontal well

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Petroleum Science 2024年第1期21卷 244-251页

作者： Ji-Lin Fan Feng Zhang Qian Chen Hao-Chen Song lu-Yu Zhong Yue-xiang Dai National Key Laboratory of Deep Oil and Gas School of GeosciencesChina University of Petroleum(East China)Qingdao 266580ShandongChina China National Logging Corporation Xi'an 710076ShaanxiChina Sinopec matrix Corporation Qingdao 266001ShandongChina

Cement density monitoring plays a vital role in evaluating the quality of cementing projects,which is of great significance to the development of oil and ***,the presence of inhomogeneous cement distribution and casing eccentricity in horizontal wells often complicates the accurate evaluation of cement azimuthal *** this regard,this paper proposes an algorithm to calculate the cement azimuthal density in horizontal wells using a multi-detector gamma-ray detection *** spatial dynamic response functions are simulated to obtain the influence of cement density on gamma-ray counts by the perturbation theory,and the contribution of cement density in six sectors to the gamma-ray recorded by different detectors is obtained by integrating the spatial dynamic response *** with the relationship between gamma-ray counts and cement density,a multi-parameter calculation equation system is established,and the regularized Newton iteration method is employed to invert casing eccentricity and cement azimuthal *** approach ensures the stability of the inversion process while simultaneously achieving an accuracy of 0.05 g/cm^(3) for the cement azimuthal *** accuracy level is ten times higher compared to density accuracy calculated using calibration ***,this algorithm enhances the accuracy of cement azimuthal density evaluation,provides valuable technical support for the monitoring of cement azimuthal density in the oil and gas industry.

关键词： Cement azimuthal density Perturbation theory Casing eccentricity Spatial dynamic response function Regularized Newton iteration

Enhancing corrosion resistance of plasma electrolytic oxidation coatings on AM50 Mg alloy by inhibitor containing Ba(NO_(3))_(2) solutions

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9期31卷 2048-2061页

作者： Jirui ma Xiaopeng lu Santosh Prasad Sah Qianqian Chen You Zhang Fuhui Wang Corrosion and Protection Center Northeastern UniversityShenyang110819China Faculty of Engineering Hokkaido UniversityHokkaido 060-8628Japan College of New materials and Chemical Engineering Beijing Institute of Petrochemical TechnologyBeijing 102617China

To enhance the long-term corrosion resistance of the plasma electrolytic oxidation(PEO)coating on the magnesium(Mg)alloy,an inorganic salt combined with corrosion inhibitors was used for posttreatment of the *** this study,the corrosion performance of PEO-coated AM50 Mg was significantly improved by loading sodium lauryl sulfonate(SDS)and sodium dodecyl benzene sulf-onate into Ba(NO_(3))_(2) post-sealing *** electron microscopy,X-ray photoelectron spectroscopy,X-ray diffraction,Fourier transform infrared spectrometer,and ultraviolet-visible analyses showed that the inhibitors enhanced the incorporation of BaO_(2) into PEO *** impedance showed that post-sealing in Ba(NO_(3))_(2)/SDS treatment enhanced corrosion resistance by three orders of *** total impedance value remained at 926Ω·cm^(2)after immersing in a 0.5wt%NaCl solution for 768 h.A salt spray test for 40 days did not show any obvious region of corrosion,proving excellent post-sealing by Ba(NO_(3))_(2)/SDS *** corrosion resistance of the coating was enhanced through the synergistic effect of BaO2 pore sealing and SDS adsorption.

关键词： Mg plasma electrolytic oxidation posttreatment corrosion resistance

Improving chitosan-based composite membrane by introducing a novel hybrid functional nano-hydroxyapatite with carboxymethyl cellulose and phytic acid

在线全文

学校读者我要写书评

暂无评论

Frontiers of Chemical Science and Engineering 2024年第6期18卷 1-15页

作者： Liuyun Jiang Yingjun ma Shuo Tang Yuqing Wang Yan Zhang Shengpei Su xiang Hu Jun He National&Local Joint Engineering Laboratory for New Petro-chemical materials and Fine Utilization of Resources Key Laboratory of Chemical Biology&Traditional Chinese Medicine Research(Ministry of Education)College of Chemistry and Chemical EngineeringHunan Normal UniversityChangsha 410081China State Key Laboratory Developmental Biology of Freshwater Fish School of Life ScienceHunan Normal UniversityChangsha 410081China Hunan Provincial Key Laboratory of Regional Hereditary Birth Defects Prevention and Control Changsha Hospital for Matemal&Child Health Care Aiffiliated to Hunan Normal UniversityChangsha 410081China

A functional hybrid nano-hydroxyapatite(carboxymethyl cellulose-phytic acid-n-HA,CMC-PA-n-HA)was prepared by adding CMC and *** results of Fourier transformation infrared spectra,X-ray diffraction,thermal gravimetric analysis and dispersion experiments indicated that the addition of CMC and PA affected the morphology,crystallinity and crystal size of hybrid n-HA,and CMC endowed hybrid n-HA with excellent *** electron microscope results showed that CMC-PA-n-HA nanoparticle could be uniformly dispersed in chitosan(CS)matrix to obtain composite membrane by casting technology,so that the highest tensile strength of CMC-PA-n-HA/CS composite membrane was 69.64%and 144.45%higher than that of CS membrane and n-HA/CS composite membrane,*** angle test showed that CMC-PA-n-HA effectively improved hydrophilicity of the CS *** simulated body fluid immersion results indicated that the CMC-PA-n-HA/CS composite membrane not only exhibited good degradability but also promoted bone-like apatite *** cell proliferation experiments proved that the introduction of PA made the composite membrane have better cell adhesion and proliferation *** tests demonstrated that PA could effectively improve the antibacterial properties of the composite membrane,which is expected to be applied as guide bone tissue regeneration membrane.

关键词： nano-hydroxyapatite chitosan dispersion guide bone tissue regeneration membrane

The Non-inclusion Diagnosability of Hypercubes Under the PMC Model

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 the Operations Research Society of China 2024年第2期12卷 478-484页

作者： Mei-Jie ma Min Xu Tong-Tong Ding xiang-Jun Li Qiang Zhu School of mathematics and Statistics Qilu University of Technology(Shandong Academy of Sciences)Jinan 250353ShandongChina School of mathematical Sciences Beijing Normal UniversityLaboratory of Mathematics and Complex SystemsMinistry of EducationBeijing 100875China School of Information and mathematics Yangtze UniversityJingzhou 434023HubeiChina School of mathematics and Statistics XiDian UniversityXi'an 710071ShaanXiChina

Diagnosability of a multiprocessor system is an important measure of the reliability of interconnection ***-level diagnosis is a primary strategy to identify the faulty processors in a multiprocessor *** on a sound assumption proposed by Zhu et ***,we proposed a new diagnosability named non-inclusion diagnosability and showed that the non-inclusion diagnosability t_(N)(Q_(n))of the hypercube under the PMC model is *** is,assume that if two vertex sets F_(1) and F_(2) are both consistent with a syndrome and F_(1)C F_(2),then F_(2) is not the faulty set which we are looking for;the faulty set F is 1-step diagnosable if|F|≤2n-2 in Q_(n) under the PMC model.

关键词： Interconnection network Diagnosability Hypercube PMC model

Data driven reduced modeling for fluidized bed with immersed tubes based on PCA and Bi-LSTM neural networksAuthor links open overlay panel

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Particuology 2024年第8期91卷 1-18页

作者： Jiabin Fang Wenkai Cu Huang Liu Huixin Zhang Hanqing Liu Jinjia Wei xiang ma Nan Zheng School of Chemical Engineering and Technology Xi'an Jiaotong UniversityXi'anShaanxi 710049China

The fast and accurate reduced-order modeling of fluidized beds is a challenging task in the field of fluid dynamics,owing to their high dimensionality and nonlinear dynamic *** this study,a nonintrusive reduced order modeling method,the reduced order model based on principal component analysis and bidirectional long short-term memory networks(PBLSTM ROM),was developed to capture complex spatio-temporal dynamics of fluidized *** combining principal component analysis and Bidirectional long-short-term memory networks,the PBLSTM ROM effectively extracted dynamic evolution information without any prior knowledge of governing equations,enabling reduced-order modeling of unsteady flow *** PBLSTM ROM was validated using the solid volume fraction and gas velocity flow fields of a fluidized bed with immersed tubes,showing superior performance over both the PLSTM and PANN ROMs in accurately capturing temporal changes in the fluidization fields,especially in the region near immersed tubes where severe fluctuations ***,the PBLSTM ROM improved the simulation speed by five orders of magnitude compared to traditional computational fluid dynamics *** findings suggest that the PBLSTM ROM presents a promising approach for analyzing the complex fluid flows in engineering practice.

关键词： Reduced order modeling Fluidized bed Deep learning Bi-LSTM

Impact of High Sodium Diet on Neovascularization and Osseointegration around Titanium Implant:An in Vivo and in Vitro Study

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Biomedical and Environmental Sciences 2024年第7期37卷 739-753页

作者： Keyuan Xu Xiaoting Tang Yun xiang Yiding Shen Zhennan Deng Pingping ma Xinkun Shen Department of Orthopaedics The Third Affiliated Hospital of Wenzhou Medical University(Ruian People’s Hospital)Wenzhou 325016ZhejiangChina School and Hospital of Stomatology Wenzhou Medical UniversityWenzhou 325027ZhejiangChina

Objective A high sodium(HS)diet is believed to affect bone metabolism *** its impact on osseointegration of titanium(Ti)implants holds significant implications for postoperative dietary management of implanted *** This investigation probed the impact of sodium ions(Na^(+))on neovascularization and osteogenesis around Ti implants in vivo,utilizing micro-computed tomography,hematoxylin and eosin staining,and immunohistochemical ***,in vitro experiments assessed the effects of varied Na^(+)concentrations and exposure durations on human umbilical vein endothelial cells(HUVECs)and MC3T3-E1 *** In vivo,increased dietary sodium(0.8%-6.0%)led to a substantial decline in CD34 positive HUVECs and new bone formation around Ti implants,alongside an increase in inflammatory *** vitro,an increase in Na^(+)concentration(140-150 mmol/L)adversely affected the proliferation,angiogenesis,and migration of HUVECs,especially with prolonged *** MC3T3-E1 cells initially exhibited less susceptibility to high Na^(+)concentrations compared to HUVECs during short-term exposure,prolonged exposure to a HS environment progressively diminished their proliferation,differentiation,and osteogenic *** These findings suggest that HS diet had a negative effect on the early osseointegration of Ti implants by interfering with the process of postoperative vascularized bone regeneration.

关键词： High-sodium Implants Vascularization Osseointegration

Multivariate statistical and bioinformatic analyses for the seasonal variations of actinobacterial community structures in a drinking water reservoir

在线全文

学校读者我要写书评

暂无评论

Journal of Environmental Sciences 2024年第3期137卷 1-17页

作者： Haihan Zhang Sixuan Pan Benma Tinglin Huang Dmitry B.Kosolapov manli ma xiang Liu Huan Liu Xiaoyan Liu Shaanxi Key Laboratory of Environmental Engineering Key Laboratory of Northwest Water ResourceEnvironment and EcologyMOEXi’an University of Architecture and TechnologyXi’an 710055China School of Environmental and Municipal Engineering Xi’an University of Architecture and TechnologyXi’an 710055China Papanin Institute for Biology of Inland Waters of Russian Academy of Sciences(IBIW RAS) 109BorokNekouzYaroslavl152742Russia

Actinobacterial community is a conspicuous part of aquatic ecosystems and displays an important role in the case of biogeochemical cycle,but little is known about the seasonal variation of actinobacterial community in reservoir ecological *** this study,the high-throughput techniques were used to investigate the structure of the aquatic actinobacterial community and its inducing water quality parameters in different *** results showed that the highest diversity and abundance of actinobacterial community occurred in winter,with Sporichthya(45.42%)being the most abundant genus and Rhodococcus sp.(29.32%)being the most abundant *** analysis and correlation analysis suggested that in autumn the dynamics of actinobacterial community were infuenced by more factors and Nocardioides ***2R5S2 was the potential keystone species which was negatively correlated with temperature(R=-0.72,Plucidated that the combined effects of DO and temperature not only in the diversity of actinobacterial community but also in other water qualities,while the physiochemical parameters(path coefficient=1.571,Pmanagement and protection in water sources.

关键词： Source water reservoir High-throughput sequencing Correlation analysis Network analysis Partial least squares path modeling

Single-source-precursor synthesis and air-plasma ablation behavior of(Ti,Zr,Hf)C/SiC ceramic nanocomposites at 2200℃

在线全文

学校读者我要写书评

暂无评论

Journal of Advanced Ceramics 2024年第7期13卷 1043-1059页

作者： Li lu Qingbo Wen Jinrun Hu Tianxing Jiang xiangchao Ren Yalei Wang Yi Zeng xiang Xiong State Key Laboratory of Powder Metallurgy Central South UniversityChangsha 410083China

Dense monolithic(Ti,Zr,Hf)C/SiC ceramic nanocomposites with four different molar ratios of metallic elements in the(Ti,Zr,Hf)C phase(i.e.,Ti:Zr:Hf=1:1:1,2:3:5,2:3:3,and 1:2:1)were prepared upon pyrolysis of novel(Ti,Zr,Hf)-containing single-source precursors(SSPs),followed by spark plasma sintering(SPS).A thorough characterization was conducted to elucidate the synthesis of the SSPs,polymer-to-ceramic transformation,chemical/phase compositions,and microstructure of the SiTiZrHfC-based *** results revealed the feasibility of synthesizing nanocomposites with high(Ti,Zr,Hf)C contents using the SSP *** nanocomposites were characterized by a unique microstructure with in situ generated(Ti,Zr,Hf)C@C core-shell nanoparticles homogeneously mixed withβ-*** ablation behavior of the nanocomposites was evaluated on an air-plasma device for 60 ***,the nanocomposites exhibited excellent ablation resistance,and the lowest linear ablation rate reached−0.58μm/s at 2200°***,the ablation resistance can be dramatically improved by precisely tailoring the atomic ratios of metal elements within the(Ti,Zr,Hf)C phase via the molecular design of the *** formation of a multiple-oxide layer with both a high-meltingpoint phase((Ti,Zr,Hf)O_(2))and low-melting-point phases((Zr,Hf)TiO4)and glassy SiO_(2),as well as their structure,played a critical role in the enhanced ablation *** uniform distribution of the high-melting-point(Ti,Zr,Hf)O_(2)nano/microparticles throughout the glassy SiO_(2)matrix significantly enhanced the viscosity and stability of the oxide layer by the pinning effect,offering superior protection against the ingress of oxygen atoms and excellent resistance to mechanical erosion.

关键词： ultrahigh-temperature ceramics(UHTCs) multicomponent carbide nanocomposites ablation behavior multiple oxides