期刊文献+
共找到1,175篇文章
< 1 2 59 >
每页显示 20 50 100
Review and Analysis of Key Techniques in Marine Sediment Sampling 被引量:5
1
作者 Shudong He Youduo Peng +2 位作者 Yongping Jin Buyan Wan Guangping Liu 《Chinese Journal of Mechanical Engineering》 SCIE EI CAS CSCD 2020年第5期150-166,共17页
Deep-sea sediment is extremely important in marine scientific research,such as that concerning marine geology and microbial communities.The research findings are closely related to the in-situ information of the sedim... Deep-sea sediment is extremely important in marine scientific research,such as that concerning marine geology and microbial communities.The research findings are closely related to the in-situ information of the sediment.One prerequisite for investigations of deep-sea sediment is providing sampling techniques capable of preventing distortion during recovery.As the fruit of such sampling techniques,samplers designed for obtaining sediment have become indispensable equipment,owing to their low cost,light weight,compactness,easy operation,and high adaptability to sea conditions.This paper introduces the research and application of typical deep-sea sediment samplers.Then,a representative sampler recently developed in China is analyzed.On this basis,a review and analysis is conducted regarding the key techniques of various deep-sea sediment samplers,including sealing,pressure and temperature retaining,low-disturbance sampling,and no-pressure drop transfer.Then,the shortcomings in the key techniques for deep-sea sediment sampling are identified.Finally,prospects for the future development of key techniques for deep-sea sediment sampling are proposed,from the perspectives of structural diversification,functional integration,intelligent operation,and high-fidelity samples.This paper summarizes the existing samplers in the context of the key techniques mentioned above,and can provide reference for the optimized design of samplers and development of key sampling techniques. 展开更多
关键词 samplER Sealing technique Pressure&temperature-retaining technique Low-disturbance sampling technique No-pressure drop transfer technique
下载PDF
Accumulator-Based Deep-Sea Microbe Gastight Sampling Technique 被引量:1
2
作者 黄中华 刘少军 金波 《China Ocean Engineering》 SCIE EI 2006年第2期335-342,共8页
The accumulator is used as a pressure compensation device to realize deep-sea microbe gastight sampling. Four key states of the accumulator are proposed to describe the pressure compensation process and a correspondin... The accumulator is used as a pressure compensation device to realize deep-sea microbe gastight sampling. Four key states of the accumulator are proposed to describe the pressure compensation process and a corresponding mathematical model is established to investigate the relationship between the results of pressure compensation and the parameters of the accumulator. Simulation results show that during the falling process of the sampler, the accumulator' s real opening pressure is greater than its precharge pressure; when the sampling depth is 6000 m and the accumulator' s precharge pressure is less than 30 MPa, to increase the accumulator' s precharge pressure can improve pressure compensation results obviously. Laboratory experiments at 60 MPa show that the acctunulator is an effective and reliable pressure compensation device for deep-sea microbe samplers, The success in sea trial at a depth of 2000 m in the South China Sea shows that the mathematical model and laboratory experiment results are reliable. 展开更多
关键词 gastight sampling technique deep-sea microbe ACCUMULATOR
下载PDF
Study on Sampling Technique of Atmospheric Radionuclide 被引量:1
3
作者 Wu Shan Liu Longbo +7 位作者 Zhang Changyun Tang Hanbing Chen Liyun Wu Yanmin Xie Feng Li Zhiming Shi Quanlin Dang Haijun 《Meteorological and Environmental Research》 CAS 2014年第6期1-3,8,共4页
Atmospheric radionuclide monitoring usually includes two sampling techniques, namely ultra-high volume aerosol samplers to collect at- mospheric particles by using filter media, and radioactive noble gas samplers to c... Atmospheric radionuclide monitoring usually includes two sampling techniques, namely ultra-high volume aerosol samplers to collect at- mospheric particles by using filter media, and radioactive noble gas samplers to collect atmospheric noble gas based on adsorption method. Atmos- pheric sampling techniques have been researched in Northwest Institute of Nuclear Technology since the Comprehensive Nuclear-Test-Ban Treaty (CTBT) was signed in 1996. Several ultra-high volume aerosol samplers and some types of radioactive xenon isotopes samplers had been devel- oped. For the aerosol sampler, the sampling flow is between 450 and 800 m3/h, with the minimum detectable concentration (MDC) of 131I less than 5 pBq/m3. For the xenon sampler, the sampling capacity of xenon is more than 4 ml per day, with MDC of l=Xe less than 0.25 mBq/m3. After the nuclear accident of Fukushima in 2011, monitoring of the atmospheric radionuclide was carried out for 3 months at Xi'an, and part of radionuclide was detected with concentrations hiaher than their backorounds in the period, includina 131I.134Cs. 137Cs and 133Xe. 展开更多
关键词 Atmospheric radionudide sampling technique Fukushima nuclear accident Aerosol sampler Xenon sampler China
下载PDF
Standardization of Leaf Sampling Technique in Jackfruit Nutrient Status Diagnosis 被引量:1
4
作者 Yan Sun Jianfeng Yang +3 位作者 Hua Wang Chao Zu Lehe Tan Gang Wu 《Agricultural Sciences》 2015年第2期232-237,共6页
In China, little information is known about the nutrient requirements of jackfruits and the traditional nutrient management usually depends on the experience. Therefore, in this study, an attempt was made to standardi... In China, little information is known about the nutrient requirements of jackfruits and the traditional nutrient management usually depends on the experience. Therefore, in this study, an attempt was made to standardize the leaf sampling technique and the suitable range of leaf nutrient concentrations for jackfruit (Artocarpus heterophyllus Lam.) nutrient status diagnosis. The sampling result was affected by canopy height, leaf age and time of sampling. Therefore, the three factors were studied. The results illustrated that the stability in level of nutrient concentrations was in 3 - 6 month-old leaves from the central part of the canopy. The most stable period was from April to May for leaf sampling. It was recommended that the stable intracanopy and stable period of nutrient concentrations could be used as the standards of leaf sampling technique. Based on the leaf sampling technique, the standard of leaf nutrient concentrations was summarized, and could be used as the standard of nutrient suitability evaluation. 展开更多
关键词 JACKFRUIT NUTRIENT STATUS DIAGNOSIS LEAF sampling technique STANDARDIZATION
下载PDF
Bias Correction Technique for Estimating Quantiles of Finite Populations under Simple Random Sampling without Replacement
5
作者 Nicholas Makumi Romanus Odhiambo Otieno +2 位作者 George Otieno Orwa Festus Were Habineza Alexis 《Open Journal of Statistics》 2021年第5期854-869,共16页
In this paper, the problem of nonparametric estimation of finite population quantile function using multiplicative bias correction technique is considered. A robust estimator of the finite population quantile function... In this paper, the problem of nonparametric estimation of finite population quantile function using multiplicative bias correction technique is considered. A robust estimator of the finite population quantile function based on multiplicative bias correction is derived with the aid of a super population model. Most studies have concentrated on kernel smoothers in the estimation of regression functions. This technique has also been applied to various methods of non-parametric estimation of the finite population quantile already under review. A major problem with the use of nonparametric kernel-based regression over a finite interval, such as the estimation of finite population quantities, is bias at boundary points. By correcting the boundary problems associated with previous model-based estimators, the multiplicative bias corrected estimator produced better results in estimating the finite population quantile function. Furthermore, the asymptotic behavior of the proposed estimators </span><span style="font-family:Verdana;">is</span><span style="font-family:Verdana;"> presented</span><span style="font-family:Verdana;">. </span><span style="font-family:Verdana;">It is observed that the estimator is asymptotically unbiased and statistically consistent when certain conditions are satisfied. The simulation results show that the suggested estimator is quite well in terms of relative bias, mean squared error, and relative root mean error. As a result, the multiplicative bias corrected estimator is strongly suggested for survey sampling estimation of the finite population quantile function. 展开更多
关键词 Quantile Function Kernel Estimator Multiplicative Bias Correction technique Simple Random sampling without Replacement
下载PDF
Upgraded Analytical Protocols in Bauxite Refining Industry Using Composite Sampling Approach to Minimize Laboratory Analysis Load
6
作者 Faisal Mubarak AlDossari 《Materials Sciences and Applications》 CAS 2023年第2期94-116,共23页
The laboratories in the bauxite processing industry are always under a heavy workload of sample collection, analysis, and compilation of the results. After size reduction from grinding mills, the samples of bauxite ar... The laboratories in the bauxite processing industry are always under a heavy workload of sample collection, analysis, and compilation of the results. After size reduction from grinding mills, the samples of bauxite are collected after intervals of 3 to 4 hours. Large bauxite processing industries producing 1 million tons of pure aluminium can have three grinding mills. Thus, the total number of samples to be tested in one day reaches a figure of 18 to 24. The sample of bauxite ore coming from the grinding mill is tested for its particle size and composition. For testing the composition, the bauxite ore sample is first prepared by fusing it with X-ray flux. Then the sample is sent for X-ray fluorescence analysis. Afterwards, the crucibles are washed in ultrasonic baths to be used for the next testing. The whole procedure takes about 2 - 3 hours. With a large number of samples reaching the laboratory, the chances of error in composition analysis increase. In this study, we have used a composite sampling methodology to reduce the number of samples reaching the laboratory without compromising their validity. The results of the average composition of fifteen samples were measured against composite samples. The mean of difference was calculated. The standard deviation and paired t-test values were evaluated against predetermined critical values obtained using a two-tailed test. It was found from the results that paired test-t values were much lower than the critical values thus validating the composition attained through composite sampling. The composite sampling approach not only reduced the number of samples but also the chemicals used in the laboratory. The objective of improved analytical protocol to reduce the number of samples reaching the laboratory was successfully achieved without compromising the quality of analytical results. 展开更多
关键词 Composite sampling Analytical Methods sampling technique Bauxite Composite sampling sampling Protocols Aluminum Ore sampling technique sampling Methods
下载PDF
A Novel Sample Introduction Technique for the Simultaneous Determination of As,Se,Ge and Hg in Chinese Medicinal Material 被引量:1
7
《Chemical Research in Chinese Universities》 SCIE CAS CSCD 2001年第4期400-406,共7页
A novel technique of Moveable Reduction Bed Hydride Generator(MRBHG)was applied tohe hydride generation or cold vapor generation of As,Se,Ge,and Hg existing In TraditionalChinese Medicinal Material(TCM).The si... A novel technique of Moveable Reduction Bed Hydride Generator(MRBHG)was applied tohe hydride generation or cold vapor generation of As,Se,Ge,and Hg existing In TraditionalChinese Medicinal Material(TCM).The simultaneous determination of the multi-elements wasperformed with ICP-MS.A solid reduction system involving the use of potassiumtetraborohydride and tartaric acid was applied to generating metal hydride or cold vaporefficiently.The factors affecting the metal cold vapor generation were studied.The mainadvantage of the technique is that only a 4μL volume of sample was required for the cold vapor 展开更多
关键词 for the Simultaneous Determination of As Se Ge A Novel sample Introduction technique CHINESE MATERIAL MEDICINAL and HG Ge
下载PDF
H_∞ Design for Sampled-Data Systems via Lifting Technique: Conditions and Limitation 被引量:5
8
作者 WANG Guang-Xiong LIU Yan-Wen HE Zhen 《自动化学报》 EI CSCD 北大核心 2006年第5期791-795,共5页
The initial motivation of the lifting technique is to solve the H∞control problems. However, the conventional weighted H∞design does not meet the conditions required by lifting, so the result often leads to a misjud... The initial motivation of the lifting technique is to solve the H∞control problems. However, the conventional weighted H∞design does not meet the conditions required by lifting, so the result often leads to a misjudgement of the design. Two conditions required by using the lifting technique are presented based on the basic formulae of the lifting. It is pointed out that only the H∞disturbance attenuation problem with no weighting functions can meet these conditions, hence, the application of the lifting technique is quite limited. 展开更多
关键词 sampled-data control system lifting technique design limitation H∞synthesis
下载PDF
Rapid Sample Product Trial Manufacture Technique for Developing Water Meter New Products 被引量:1
9
作者 YAO Ling WANG Kai-tuo XU Liang 《International Journal of Plant Engineering and Management》 2008年第4期225-231,共7页
This paper advances the viewpoints and methods of the rapid sample product trial manufacture technique for developing water meter new products by CAD and simulation, computer virtual assembling and optimizing, rapid m... This paper advances the viewpoints and methods of the rapid sample product trial manufacture technique for developing water meter new products by CAD and simulation, computer virtual assembling and optimizing, rapid machining process and measurement etc. as the design and sample product trial manufacture process of water meter new products are long in product development period, and low in product development efficiency in the present time. 展开更多
关键词 water meter design sample product trial manufacture rapid sample product trial manufacture technique
下载PDF
Over-sampling algorithm for imbalanced data classification 被引量:9
10
作者 XU Xiaolong CHEN Wen SUN Yanfei 《Journal of Systems Engineering and Electronics》 SCIE EI CSCD 2019年第6期1182-1191,共10页
For imbalanced datasets, the focus of classification is to identify samples of the minority class. The performance of current data mining algorithms is not good enough for processing imbalanced datasets. The synthetic... For imbalanced datasets, the focus of classification is to identify samples of the minority class. The performance of current data mining algorithms is not good enough for processing imbalanced datasets. The synthetic minority over-sampling technique(SMOTE) is specifically designed for learning from imbalanced datasets, generating synthetic minority class examples by interpolating between minority class examples nearby. However, the SMOTE encounters the overgeneralization problem. The densitybased spatial clustering of applications with noise(DBSCAN) is not rigorous when dealing with the samples near the borderline.We optimize the DBSCAN algorithm for this problem to make clustering more reasonable. This paper integrates the optimized DBSCAN and SMOTE, and proposes a density-based synthetic minority over-sampling technique(DSMOTE). First, the optimized DBSCAN is used to divide the samples of the minority class into three groups, including core samples, borderline samples and noise samples, and then the noise samples of minority class is removed to synthesize more effective samples. In order to make full use of the information of core samples and borderline samples,different strategies are used to over-sample core samples and borderline samples. Experiments show that DSMOTE can achieve better results compared with SMOTE and Borderline-SMOTE in terms of precision, recall and F-value. 展开更多
关键词 imbalanced data density-based spatial clustering of applications with noise(DBSCAN) synthetic minority over sampling technique(SMOTE) over-sampling.
下载PDF
Real-time prediction of rock mass classification based on TBM operation big data and stacking technique of ensemble learning 被引量:21
11
作者 Shaokang Hou Yaoru Liu Qiang Yang 《Journal of Rock Mechanics and Geotechnical Engineering》 SCIE CSCD 2022年第1期123-143,共21页
Real-time prediction of the rock mass class in front of the tunnel face is essential for the adaptive adjustment of tunnel boring machines(TBMs).During the TBM tunnelling process,a large number of operation data are g... Real-time prediction of the rock mass class in front of the tunnel face is essential for the adaptive adjustment of tunnel boring machines(TBMs).During the TBM tunnelling process,a large number of operation data are generated,reflecting the interaction between the TBM system and surrounding rock,and these data can be used to evaluate the rock mass quality.This study proposed a stacking ensemble classifier for the real-time prediction of the rock mass classification using TBM operation data.Based on the Songhua River water conveyance project,a total of 7538 TBM tunnelling cycles and the corresponding rock mass classes are obtained after data preprocessing.Then,through the tree-based feature selection method,10 key TBM operation parameters are selected,and the mean values of the 10 selected features in the stable phase after removing outliers are calculated as the inputs of classifiers.The preprocessed data are randomly divided into the training set(90%)and test set(10%)using simple random sampling.Besides stacking ensemble classifier,seven individual classifiers are established as the comparison.These classifiers include support vector machine(SVM),k-nearest neighbors(KNN),random forest(RF),gradient boosting decision tree(GBDT),decision tree(DT),logistic regression(LR)and multilayer perceptron(MLP),where the hyper-parameters of each classifier are optimised using the grid search method.The prediction results show that the stacking ensemble classifier has a better performance than individual classifiers,and it shows a more powerful learning and generalisation ability for small and imbalanced samples.Additionally,a relative balance training set is obtained by the synthetic minority oversampling technique(SMOTE),and the influence of sample imbalance on the prediction performance is discussed. 展开更多
关键词 Tunnel boring machine(TBM)operation data Rock mass classification Stacking ensemble learning sample imbalance Synthetic minority oversampling technique(SMOTE)
下载PDF
Research Progress on Analysis and Detection Techniques of Veterinary Drug Residues in Animal Foods 被引量:1
12
作者 Bing LI Xuzheng ZHOU +3 位作者 Fusheng CHENG Xiaojuan WEI Weiwei Wang Jiyu ZHANG 《Agricultural Biotechnology》 CAS 2019年第5期60-64,69,共6页
As an important guarantee for the prevention and control of animal diseases,veterinary drugs have important functions in improving animal production performance and product quality and maintaining ecological balance.T... As an important guarantee for the prevention and control of animal diseases,veterinary drugs have important functions in improving animal production performance and product quality and maintaining ecological balance.They are an important guarantee for the healthy development of animal husbandry,food safety and public health.However,the irrational use and abuse of veterinary drugs and feed pharmaceutical additives are widespread,causing harmful substances in animal foods and damage to human health,and threatening the sustainable development of the environment and animal husbandry as well.In order to ensure human health,it is urgent to develop a simple,rapid,high-sensitivity,high-throughput and low-cost veterinary drug residue detection technology.In this paper,the sample pretreatment methods and detection techniques for the analysis of veterinary drug residues in animal foods were reviewed. 展开更多
关键词 ANIMAL FOOD VETERINARY DRUG RESIDUE samplE preparation Detection techniques
下载PDF
Comparative Study of Two Vitreous Humor Sampling Methods in Rabbits
13
作者 汪岚 周伟光 +2 位作者 任亮 刘茜 刘良 《Journal of Huazhong University of Science and Technology(Medical Sciences)》 SCIE CAS 2006年第4期475-477,共3页
Summary: To compare and evaluate two methodologies, entire-sampling and micro-sampling for the harvesting of vitreous humor, the vitreous humor of rabbits were sampled with the two methods respectively, and the conce... Summary: To compare and evaluate two methodologies, entire-sampling and micro-sampling for the harvesting of vitreous humor, the vitreous humor of rabbits were sampled with the two methods respectively, and the concentrations of calcium, chlorine, potassium, sodium and phosphorus of the were measured. The results showed that the differences in the variance coefficient and two-eye concentrations of micro-sampled specimens were less than those of the entire-sampled specimens. In the micro-sampling group, the concentrations of repeated micro-sampling showed no differences among different groups (P〉0.05) and the intra-ocular fluid dynamics did not have significant influence on post-mortem sampling. The sampling technique may affect the concentrations of specimen collected. Our study suggests that micro-sampling is less influenced by the human factor and is reliable, reproducible, and more suitable for forensic investigation. 展开更多
关键词 vitreous humor sampling techniques comparative study
下载PDF
Optimization of the Sampling Method for Genetic Variation Survey in Maize Populations Using SSR Markers
14
作者 Fengcheng SUN Yong FENG +4 位作者 Zhuo YU Chen Bai Zengjuan FU Chuanzong YU Zhizhuang SHAO 《Agricultural Biotechnology》 CAS 2013年第5期39-44,共6页
In this study, single-plant DNA sampling method, multi-plant leaf-mixing DNA sampling method, and single-plant DNA-mixing sampling method were adopted, to analyze the genetic variations of Meng A population and C popu... In this study, single-plant DNA sampling method, multi-plant leaf-mixing DNA sampling method, and single-plant DNA-mixing sampling method were adopted, to analyze the genetic variations of Meng A population and C population 1 using SSR markers and establish the optimal technological system for analyzing genetic diversity of maize populations. DNA samples in different treatments were amplified using 34 SSR primers which were uniformly distributed in ten chromosomes of maize. Polyacrylamidc gel electrophoresis was performed to analyze the polymorphism information content and genetic similarity coefficient of 60 individuals and compare the numbers of alleles amplified from DNA samples in different treatments. The results indicated that Meng A population and C population 1 both had relatively abundant genetic variations and the established technological system could be applied in researches of maize genetic diversity. Extracting DNA samples from mixed leaves of 12 individuals with five replications is the best sampling method, which could achieve similar results to mixed DNA samples of 12 individuals. This sampling method can be applied to analyze the genetic relationship among a large number of maize populations, which can not only reduce the workload, but also significantly improve the efficiency. 展开更多
关键词 Maize population SSR technique Genetic diversity Optimal sampling
下载PDF
Determination of Heavy Metals at Sub-ppb Levels in Water by Graphite Furnace Atomic Absorption Spectrometry Using a Direct Introduction Technique after Preconcentration with an Iminodiacetate Extraction Disk
15
作者 Tetsuo Inui Atsuko Kosuge +4 位作者 Atsushi Ohbuchi Kazuhiro Fujita Yuya Koike Masaru Kitano Toshihiro Nakamura 《American Journal of Analytical Chemistry》 2012年第10期683-692,共10页
A direct analysis method combining an iminodiacetate extraction disk (IED) with graphite furnace atomic absorption spectrometry was developed for the determination of Co, Ni, Cu, Cd, Sn, Pb, and Bi at sub-ppb levels i... A direct analysis method combining an iminodiacetate extraction disk (IED) with graphite furnace atomic absorption spectrometry was developed for the determination of Co, Ni, Cu, Cd, Sn, Pb, and Bi at sub-ppb levels in water. A 100 mL water sample was adjusted to pH 5.6 with nitric acid and a 1 mol?L–1 ammonium acetate solution, and then passed through an IED (diameter, 47 mm;effective filtering diameter, 35 mm) at a flow rate of 80 - 100 mL?min–1 to preconcentrate seven analytes. The IED was dried at 100?C for 20 min in an electric oven, and 110 - 145 small disks, each 2 mm in diameter, were punched out from the IED. A small disk was introduced into the graphite furnace and atomized according to a heating program. For Cd, Sn, Pb, and Bi measurements, Pd was used as a chemical modifier to enhance the absorbances. Calibration was performed using aqueous standard solutions. The detection limits, corresponding to three times the standard deviation (n = 5) of the blank values, were 0.092 μg·L–1 for Co, 0.12 μg·L–1 for Ni, 0.40 μg·L–1 for Cu, 0.077 μg·L–1 for Cd, 0.92 μg·L–1 for Sn, 0.61 μg·L–1 for Pb, and 0.80 μg·L–1 for Bi with an enrichment factor of 140 using a 100-mL water sample. A spike test for the seven analytes in tap water, rainwater, river water, and mineral drinking water showed quantitative recoveries (93% - 108%). 展开更多
关键词 Water sample Heavy Metal PRECONCENTRATION IMINODIACETATE EXTRACTION DISK DIRECT INTRODUCTION technique Graphite Furnace Atomic Absorption Spectrometry
下载PDF
Determination of the full-field stress and displacement using photoelasticity and sampling moirémethod in a 3D-printed model
16
作者 Zhangyu Ren Qi Zhang +1 位作者 Yang Ju Huimin Xie 《Theoretical & Applied Mechanics Letters》 CAS CSCD 2022年第6期409-418,共10页
The quantitative characterization of the full-field stress and displacement is significant for analyzing the failure and instability of engineering materials.Various optical measurement techniques such as photoelastic... The quantitative characterization of the full-field stress and displacement is significant for analyzing the failure and instability of engineering materials.Various optical measurement techniques such as photoelasticity,moiréand digital image correlation methods have been developed to achieve this goal.However,these methods are difficult to incorporate to determine the stress and displacement fields simultaneously because the tested models must contain particles and grating for displacement measurement;however,these elements will disturb the light passing through the tested models using photoelasticity.In this study,by combining photoelasticity and the sampling moirémethod,we developed a method to determine the stress and displacement fields simultaneously in a three-dimensional(3D)-printed photoelastic model with orthogonal grating.Then,the full-field stress was determined by analyzing 10 photoelastic patterns,and the displacement fields were calculated using the sampling moirémethod.The results indicate that the developed method can simultaneously determine the stress and displacement fields. 展开更多
关键词 Stress fields Displacement fields PHOTOELASTICITY sampling moirémethod 3D printing technique
下载PDF
基于SMOTE-IKPCA-SeNet深度迁移学习的小批量生产质量预测研究 被引量:1
17
作者 杨剑锋 崔少红 +1 位作者 段家琦 王宁 《工业工程》 2024年第2期98-106,157,共10页
随着智能制造技术的发展和客户个性化需求的增加,多品种小批量生产方式逐渐成为制造业的主流。面向大批量生产、以统计过程控制为核心的质量管理方式并不适用于小批量生产。针对复杂生产过程存在参数多、非线性和交互作用的问题,提出利... 随着智能制造技术的发展和客户个性化需求的增加,多品种小批量生产方式逐渐成为制造业的主流。面向大批量生产、以统计过程控制为核心的质量管理方式并不适用于小批量生产。针对复杂生产过程存在参数多、非线性和交互作用的问题,提出利用深度迁移学习的方式将历史生产数据作为源域迁移至小样本目标产品数据进行质量预测。首先,通过合成少数类过采样技术(synthetic minority over-sampling technique,SMOTE)和改进的核主成分分析(improved kernel principal component analysis,IKPCA)算法筛选源域和目标域的可迁移特征,这不仅兼顾了特征重要性和可迁移性,还减少了“负迁移”,提高了模型泛化能力;然后,采用结合通道注意力机制的卷积神经网络SeNet构建基于深度迁移学习的质量预测模型。仿真结果表明,随着目标域样本的增加,所提方法的预测准确性明显优于广泛采用的支持向量机建模方法。同时,所提可迁移特征筛选方法显著提高了深度迁移学习的质量预测效果,为复杂的小批量生产过程质量保证提供了新方法。 展开更多
关键词 小批量生产质量预测 深度迁移学习 SMOTE IKPCA SeNet
下载PDF
基于自动重要抽样方法的减方差技巧体系构建与验证
18
作者 武祯 郝以昇 +6 位作者 浦彦恒 周扬 杲申申 邱睿 马锐垚 张辉 李君利 《原子能科学技术》 EI CAS CSCD 北大核心 2024年第3期630-637,共8页
屏蔽计算问题根据求解目标不同一般可分为源-探测器问题、区域问题和全局问题。MCShield研究团队针对3类问题中存在的深穿透问题提出了相应的减方差技巧,本文以此为基础构建了基于自动重要抽样(AIS)方法的减方差技巧体系,并开展了验证... 屏蔽计算问题根据求解目标不同一般可分为源-探测器问题、区域问题和全局问题。MCShield研究团队针对3类问题中存在的深穿透问题提出了相应的减方差技巧,本文以此为基础构建了基于自动重要抽样(AIS)方法的减方差技巧体系,并开展了验证工作。针对源-探测器问题,采用NUREG/CR-6115 PWR压力容器计算基准题对小探测器自动重要抽样(SDAIS)方法进行验证。结果表明,SDAIS方法的计算效率约为AIS方法的7倍。此外还提出并验证了基于AIS伴随蒙特卡罗的耦合减方差(AIS-CADIS)方法,将AIS方法引入到蒙特卡罗伴随计算中,取得了良好的效果。针对全局问题,提出网格化-AIS方法并使用简化反应堆屏蔽计算算例进行验证,结果表明,网格化-AIS方法的计算效率是AIS方法的12倍左右,是直接蒙特卡罗方法的290倍左右。 展开更多
关键词 蒙特卡罗 自动重要抽样方法 减方差技巧 MCShield程序
下载PDF
基于CD-BSMOTE的D-S证据融合变压器故障诊断
19
作者 鲁玲 高诚 +3 位作者 熊威 龚康 马辉 张鑫 《水电能源科学》 北大核心 2024年第5期192-196,共5页
针对变压器油中溶解气体数据集不均衡特性对故障诊断结果的影响,提出一种基于清除临界点改进的边界合成少数类过采样算法均衡数据集和Pearson冲突距离改进D-S证据融合的变压器故障诊断模型。首先,对少数类样本进行均衡化处理,根据K-mean... 针对变压器油中溶解气体数据集不均衡特性对故障诊断结果的影响,提出一种基于清除临界点改进的边界合成少数类过采样算法均衡数据集和Pearson冲突距离改进D-S证据融合的变压器故障诊断模型。首先,对少数类样本进行均衡化处理,根据K-means聚类结果清除处于临界位置的样本;其次,搭建梯度提升树、随机森林、BP神经网络的故障诊断模型,实现变压器故障初步诊断;接着引入Pearson冲突距离改进D-S证据融合模型,实现诊断结果的融合决策;最后,经实际算例分析,诊断精确率达到92.65%。结果表明,所建模型能有效解决数据不平衡对诊断结果的影响,提高故障诊断精度。 展开更多
关键词 故障诊断 油中溶解气体分析 边界合成少数类过采样 Pearson冲突距离 D-S证据融合
下载PDF
青海云杉一种重要害虫烟翅腮扁蜂越冬幼虫的空间分布型与抽样技术
20
作者 张鑫 刘超 +2 位作者 马嘉瑜 兰珍珍 王新谱 《农业科学研究》 2024年第1期56-60,73,共6页
为研究烟翅腮扁蜂(Cephalcia infumata)越冬幼虫在地下的空间分布,采用聚集度指标法、Taylor幂法则和Iwao回归法对烟翅腮扁蜂越冬幼虫的空间分布型进行分析。基于Iwao回归模型确定烟翅腮扁蜂越冬幼虫的理论数,通过序贯抽样技术得到不同... 为研究烟翅腮扁蜂(Cephalcia infumata)越冬幼虫在地下的空间分布,采用聚集度指标法、Taylor幂法则和Iwao回归法对烟翅腮扁蜂越冬幼虫的空间分布型进行分析。基于Iwao回归模型确定烟翅腮扁蜂越冬幼虫的理论数,通过序贯抽样技术得到不同允许误差(D=0.1、0.2、0.3)和经济阈值(m_0=3、5、10头/株)下的最大理论抽样数。结果表明,在宁夏罗山国家级自然保护区内,烟翅腮扁蜂1年发生1代,越冬幼虫主要分布在0~10 cm土壤中,在阳坡发生量大,呈聚集分布,符合负二项分布,种群聚集均数与种群平均密度呈线性关系,聚集程度随种群密度增大而增强。基于I_(wao)回归模型,在t=1.96时,当m_0=3头/株,D=0.1、0.2、0.3时,最大理论抽样数分别为596、149、66株;当m_0=5头/株,D=0.1、0.2、0.3时,最大理论抽样数分别为477、119、53株;当m_0=10头/株,D=0.1、0.2、0.3时,最大理论抽样数分别为388、97、43株。烟翅腮扁蜂的种群空间分布型为聚集分布,主要发生在林缘。基于序贯抽样确定的最大抽样数可用于指导烟翅腮扁蜂的监测和防治。 展开更多
关键词 烟翅腮扁蜂 越冬幼虫 空间分布 抽样技术
下载PDF
上一页 1 2 59 下一页 到第
使用帮助 返回顶部