期刊文献+
共找到31篇文章
< 1 2 >
每页显示 20 50 100
改进的U-Net网络小断层识别技术在玛湖凹陷玛中地区三叠系白碱滩组的应用 被引量:1
1
作者 宋志华 李垒 +2 位作者 雷德文 张鑫 凌勋 《岩性油气藏》 CAS CSCD 北大核心 2024年第3期40-49,共10页
利用改进的U-Net网络小断层识别技术,对准噶尔盆地玛湖凹陷玛中地区三叠系白碱滩组的小断层进行了识别。研究结果表明:(1)构造导向滤波预处理能有效改善地震资料的品质,提高断层识别的准确率。加入了跳跃连接和中继监督、正态标准化和... 利用改进的U-Net网络小断层识别技术,对准噶尔盆地玛湖凹陷玛中地区三叠系白碱滩组的小断层进行了识别。研究结果表明:(1)构造导向滤波预处理能有效改善地震资料的品质,提高断层识别的准确率。加入了跳跃连接和中继监督、正态标准化和聚焦均方损失函数的U-Net网络方法,对小断层的精细识别能力有所提升。(2)使用200组训练样本集和20组验证样本集,模型地震数据由反射系数与雷克子波褶积生成,断层由人工标注而成。选取最优的网络模型参数,并在合成的含噪地震数据上分别利用相干属性、常规U-Net网络方法及改进的U-Net网络方法进行测试,构造导向滤波有效突出了断层的边界,且增强了同相轴的横向连续性,改进后的U-Net网络方法对于7 m以上断距的断层可进行有效识别。(3)对于玛湖凹陷玛中地区三叠系白碱滩组高角度走滑断裂和伴生小断距次级断裂的识别,改进后的U-Net网络方法的识别精度明显高于相干属性和常规U-Net网络方法,研究区大侏罗沟断裂北翼的(3)号与(4)号砂体,是拓展MZ4井区三叠系白碱滩组高效勘探的有利区。 展开更多
关键词 U-Net网络 断层识别 高角度走滑断裂 伴生小断距次级断裂 正态标准化 聚焦均方损失函数 白碱滩组 三叠系 玛湖凹陷
下载PDF
夹马口灌区干支渠道输水损失考核方法探索 被引量:2
2
作者 孙璐 谢自立 《山西水利》 2012年第12期43-44,共2页
为进一步加强供水管理,夹马口灌区增设了"千方.公里输水损失"考核,阐述了考核的具体办法及实施考核的意义,考核实施以来,灌区上下游均衡受益,且大量减少了输水损失,建议将"千方.公里输水损失"作为渠道水资源单耗指... 为进一步加强供水管理,夹马口灌区增设了"千方.公里输水损失"考核,阐述了考核的具体办法及实施考核的意义,考核实施以来,灌区上下游均衡受益,且大量减少了输水损失,建议将"千方.公里输水损失"作为渠道水资源单耗指标在灌区推广。 展开更多
关键词 水的利用系数 ·公里损失 损失水量 夹马口灌区
下载PDF
夹马口灌区渠道输水实行“千方·公里损失”考核的实践
3
作者 谢自立 孙璐 《中国水利》 2013年第11期30-30,55,共2页
一、考核指标的提出 夹马口灌区是一个大型电力提黄灌区.工程分三级九站.泵站设计扬程70.225m.灌区由夹马口、小樊和泉杜灌区三部分组成.总设计灌溉面积90.9万亩(6.06万hm^2)。
关键词 灌区 水利用系数 ·公里损失 考核
下载PDF
相对损失函数下的倍度保费(英文) 被引量:6
4
作者 张佳佳 《应用概率统计》 CSCD 北大核心 2007年第2期157-164,共8页
传统的倍度保费公式利用均方损失函数估计特定保人的风险.然而,索取保费与真实保费之间的比例比它们差的绝对值更适合于衡量保费的公平性.基于这一点,我们提出了两种计算保费的损失函数:均方相对损失函数和熵相对损失函数,并且给出了... 传统的倍度保费公式利用均方损失函数估计特定保人的风险.然而,索取保费与真实保费之间的比例比它们差的绝对值更适合于衡量保费的公平性.基于这一点,我们提出了两种计算保费的损失函数:均方相对损失函数和熵相对损失函数,并且给出了倍度因子的估计公式及它们的性质. 展开更多
关键词 方损失 损失 相对损失 倍度.
下载PDF
部分先验信息Γ下分布族参数的Γ-容许估计 被引量:1
5
作者 陈兰祥 钱伟民 《同济大学学报(自然科学版)》 EI CAS CSCD 1995年第1期69-73,共5页
讨论了在某一族先验分布族Γ或者一定的先验矩限制Γ下分布参数的Γ-容许估计,给出了相应的充分条件,分别推广了文献[1]和[2]的结果,并给出了若干例子.
关键词 Г-容许性 先验分布 方损失
下载PDF
单参数指数族的一类非线性容许估计的充分条件 被引量:1
6
作者 陈欣城 窦元烈 《石家庄铁道大学学报(自然科学版)》 1989年第4期20-27,共8页
1981年D.Ralescu,和S.Ralescu,在文献[1]中,对单参数指数型分布,在均方损失下,得到了非线性估计(ax+b)/(cx+d)是g(θ)的可容许估计的充分条件。该条件复杂,验证也较麻烦。本文用Bayes方法,得到了另一个充分条件,易于验证。在N(o,σ~2)中... 1981年D.Ralescu,和S.Ralescu,在文献[1]中,对单参数指数型分布,在均方损失下,得到了非线性估计(ax+b)/(cx+d)是g(θ)的可容许估计的充分条件。该条件复杂,验证也较麻烦。本文用Bayes方法,得到了另一个充分条件,易于验证。在N(o,σ~2)中,得到了当 a】0,b】0,d】(n-2)a时,(ax+b)是σ~2的容许估计。文献[1]中,得到的估计类(x+k)/(n+a),k】0,显然包含在我们的估计类中。 展开更多
关键词 方损失 BAYES估计 容许估计
下载PDF
基于偿付能力的最优再保险策略 被引量:4
7
作者 王丽珍 李秀芳 《系统工程学报》 CSCD 北大核心 2012年第1期44-51,共8页
采用最大索赔再保费定价原则,结合VaR、CTE、TV三种风险测度方法,通过研究最小化偿付不足风险的概率、期望损失以及均方期望超额损失等再保险问题,得到相应的最优再保险策略,并结合案例对各种最优策略进行静态分析.研究发现,当偿付能力... 采用最大索赔再保费定价原则,结合VaR、CTE、TV三种风险测度方法,通过研究最小化偿付不足风险的概率、期望损失以及均方期望超额损失等再保险问题,得到相应的最优再保险策略,并结合案例对各种最优策略进行静态分析.研究发现,当偿付能力基于VaR或者CTE时,最优的再保险策略是去尾停止损失再保险,这说明原保险公司此时应该更注重对中等巨额损失的保障,而没有动力去保障极值损失;当偿付能力基于TV时,最优策略是带限额的停止损失再保险,此时,保险公司为了保证经营的稳定性,势必会将一部分极值损失分保. 展开更多
关键词 最大索赔原则 偿付能力 最优再保险 期望超额损失
下载PDF
二参数指数分布有序约束位置参数的保序估计 被引量:1
8
作者 汤银才 《上海师范大学学报(自然科学版)》 1995年第3期13-18,共6页
假定有两个来自于不同的二参数指数分布的独立样本,它们的位置参数满足顺序约束.本文分三种情形对常规估计作出了改进,给出了有约束位置参数的估计,并在一组特殊的参数下证明了它们在均方损失函数下优于常规估计.
关键词 位置参数 顺序约束 方损失函数 保序估计
下载PDF
统计预测
9
作者 张尧庭 《统计教育》 1997年第5期31-34,共4页
统计预测上海财经大学张尧庭统计预测有以下特点:(1)结论是不确定的。例如预测明天“降水”的概率是多少。它与一些确定性的预测不同,对不同的置信度相应的结论也会不同。(2)从数据和模型出发,具有客观性。它不随人的判断的不... 统计预测上海财经大学张尧庭统计预测有以下特点:(1)结论是不确定的。例如预测明天“降水”的概率是多少。它与一些确定性的预测不同,对不同的置信度相应的结论也会不同。(2)从数据和模型出发,具有客观性。它不随人的判断的不同而改变,只要依据的模型和数据是可... 展开更多
关键词 误差损失 回归 卡尔曼滤波 统计预测 预测值 统计 组合预测 线性预测 损失函数 条件期望
下载PDF
变步长BLSTM集成学习股票预测 被引量:2
10
作者 王子玥 谢维波 李斌 《华侨大学学报(自然科学版)》 CAS 北大核心 2019年第2期269-276,共8页
提出采用变步长双向长短期记忆网络(BLSTM)集成学习方法学习历史数据中股票价格变动的规律.针对股票涨跌变化的预测改进均方误差(MSE)损失函数,采用简易的模拟交易盈利评价指标以更好地度量预测模型在金融市场中的期望表现.通过前10~5... 提出采用变步长双向长短期记忆网络(BLSTM)集成学习方法学习历史数据中股票价格变动的规律.针对股票涨跌变化的预测改进均方误差(MSE)损失函数,采用简易的模拟交易盈利评价指标以更好地度量预测模型在金融市场中的期望表现.通过前10~50步长的数据训练BLSTM,预测下1min各股票的涨跌变化.实验结果验证了不同数据预处理下,改进损失函数的有效性及变步长集成方法相对于单一网络的有效性. 展开更多
关键词 双向长短期记忆网络 集成学习 变步长 股票价格 改进均误差损失
下载PDF
敷设去耦层水下简支撑板的振动特性
11
作者 赵军 何世平 《舰船科学技术》 北大核心 2013年第2期17-22,共6页
将去耦层等效为弹性元件,利用简正模式法推导出重流体作用下敷设去耦层的有限薄板在点激励下的均方速度和干面均方速度插入损失。理论分析得出,在低频区域(k0a<<1),去耦层对板的振动影响较小;在高频区域(k0a>>1),敷设去耦... 将去耦层等效为弹性元件,利用简正模式法推导出重流体作用下敷设去耦层的有限薄板在点激励下的均方速度和干面均方速度插入损失。理论分析得出,在低频区域(k0a<<1),去耦层对板的振动影响较小;在高频区域(k0a>>1),敷设去耦层后板的振动趋向于真空中的振动,因此敷设后板的振动相对未敷设的板是放大的。随着去耦层的阻尼因子增加,放大现象减弱。 展开更多
关键词 去耦覆盖层 简正模态法 速度 干面均速度插入损失
下载PDF
Inhibiting Effect of Ciprofloxacin,Norfloxacin and Ofloxacin on Corrosion of Mild Steel in Hydrochloric Acid 被引量:7
12
作者 庞雪辉 冉祥滨 +2 位作者 匡飞 解建东 侯保荣 《Chinese Journal of Chemical Engineering》 SCIE EI CAS CSCD 2010年第2期337-345,共9页
The inhibiting effect of ciprofloxacin,norfloxacin and ofloxacin on the corrosion of mild steel in 1 mol·L-1 HCl and the mechanism were studied at different temperatures using mass loss measurement,electrochemica... The inhibiting effect of ciprofloxacin,norfloxacin and ofloxacin on the corrosion of mild steel in 1 mol·L-1 HCl and the mechanism were studied at different temperatures using mass loss measurement,electrochemical method,and X-ray photoelectron spectroscopy(XPS) .Effective inhibition was shown by mass loss,potentiodynamic polarization and impedance spectroscopy measurement.The corrosion rate of the metal in the mass loss measurement,and the corrosion reaction on cathode and anode in the electrochemical measurement were accelerated when temperature was increased.XPS results showed that the inhibitors adsorbed effectively on the metal surface. 展开更多
关键词 CORROSION INHIBITION electrochemical impedance spectroscopy potentiodynamic polarization X-ray photoelectron spectroscopy
下载PDF
A polynomial smooth epsilon-support vector regression based on cubic spline interpolation
13
作者 任斌 He Chunhong +2 位作者 Liu Huijie Yang Lei Xie Guobo 《High Technology Letters》 EI CAS 2014年第2期187-194,共8页
Regression analysis is often formulated as an optimization problem with squared loss functions. Facing the challenge of the selection of the proper function class with polynomial smooth techniques applied to support v... Regression analysis is often formulated as an optimization problem with squared loss functions. Facing the challenge of the selection of the proper function class with polynomial smooth techniques applied to support vector regression models, this study takes cubic spline interpolation to generate a new polynomial smooth function |×|ε^ 2, in g-insensitive support vector regression. Theoretical analysis shows that Sε^2 -function is better than pε^2 -function in properties, and the approximation accuracy of the proposed smoothing function is two order higher than that of classical pε^2 -function. The experimental data shows the efficiency of the new approach. 展开更多
关键词 support vector regression ε-insensitive loss function SMOOTH polynomial function cubic spline interpolation
下载PDF
Ammonia Volatilization from Urea Applied to Acid Paddy Soil in Southern China and Its Control 被引量:12
14
作者 CAIGUI-XIN PENGGUANG-HAO 《Pedosphere》 SCIE CAS CSCD 1992年第4期345-354,共10页
Results showed that ammonia loss from urea broadcast into floodwater and incorporated into soil at transplanting was as high as 40% of applied N,and the corresponding total nitrogen (N) loss was 56%.Ammonia loss was m... Results showed that ammonia loss from urea broadcast into floodwater and incorporated into soil at transplanting was as high as 40% of applied N,and the corresponding total nitrogen (N) loss was 56%.Ammonia loss was measured with simplified micrometeorological method (ammonia sampler),and total N loss was concurrently measured using ^15N balance technique.The experiment was conducted under strong sunshine conditions on acid paddy soil derived from Quaternary red clay.The ammonia loss in this particular condition was much greater than those obtained from previous studies when urea was also applied to acid paddy soil but under cloudy conditions.It is concluded that the strong sunshine conditions with high temperature and shallow floodwater during the period of present experiment favoured ammonia volatilization.Application of stearyl alcohol on the surface of the floodwater reduced ammonia loss to 23% of applied N.However,the effect of stearyl alcohol was short-lived,probably due to the microbiological decomposition. 展开更多
关键词 ammonia volatilization flooded rice N loss surface film UREA
下载PDF
Evapotranspiration, Yield and Crop Coefficient of Irrigated Maize Under Straw Mulch 被引量:12
15
作者 ZHANG Xi-Ying CHEN Su-Ying PEI Dong LIU Meng-Yu SUN Hong-Yong 《Pedosphere》 SCIE CAS CSCD 2005年第5期576-584,共9页
Maize (Zea mays L.), a staple crop grown from June to September during the rainy season on the North China Plain,is usually inter-planted in winter wheat (Triticum aestivum L.) fields about one week before harvesting ... Maize (Zea mays L.), a staple crop grown from June to September during the rainy season on the North China Plain,is usually inter-planted in winter wheat (Triticum aestivum L.) fields about one week before harvesting of the winterwheat. In order to improve irrigation efficiency in this region of serious water shortage, field studies in 1999 and 2001, twodry seasons with less than average seasonal rainfall, were conducted with up to five irrigation applications to determineevapotranspiration, calculate the crop coefficient, and optimize the irrigation schedule with maize under mulch, as well asto establish the effects of irrigation timing and the number of applications on grain yield and water use efficiency (WUE)of maize. Results showed that with grain production at about 8 000 kg ha-1 the total evapotranspiration and WUE ofirrigated maize under mulch were about 380-400 mm and 2.0-2.2 kg m-3, respectively. Also in 2001 WUE of maizewith mulch for the treatment with three irrigations was 11.8% better than that without mulch. In the 1999 and 2001seasons, maize yield significantly improved (P = 0.05) with four irrigation applications, however, further increases werenot significant. At the same time there were no significant differences for WUE with two to four irrigation applications.In the 2001 season mulch lead to a decrease of 50 mm in the total soil evaporation, and the maize crop coefficient undermulch varied between 0.3-1.3 with a seasonal average of 1.0. 展开更多
关键词 crop coefficient MAIZE MULCH water use efficiency
下载PDF
Solving Nonlinear Differential Equation Governing on the Rigid Beams on Viscoelastic Foundation by AGM 被引量:1
16
作者 M. R. Akbari D. D. Ganji +1 位作者 A. K. Rostami M. Nimafar 《Journal of Marine Science and Application》 CSCD 2015年第1期30-38,共9页
In the present paper a vibrational differential equation governing on a rigid beam on viscoelastic foundation has been investigated. The nonlinear differential equation governing on this vibrating system is solved by ... In the present paper a vibrational differential equation governing on a rigid beam on viscoelastic foundation has been investigated. The nonlinear differential equation governing on this vibrating system is solved by a simple and innovative approach, which has been called Akbari-Ganji's method (AGM). AGM is a very suitable computational process and is usable for solving various nonlinear differential equations. Moreover, using AGM which solving a set of algebraic equations, complicated nonlinear equations can easily be solved without any mathematical operations. Also, the damping ratio and energy lost per cycle for three cycles have been investigated. Furthermore, comparisons have been made between the obtained results by numerical method (Runk45) and AGM. Results showed the high accuracy of AGM. The results also showed that by increasing the amount of initial amplitude of vibration (A), the value of damping ratio will be increased, and the energy lost per cycle decreases by increasing the number of cycle. It is concluded that AGM is a reliable and precise approach for solving differential equations. On the other hand, it is better to say that AGM is able to solve linear and nonlinear differential equations directly in most of the situations. This means that the final solution can be obtained without any dimensionless procedure Therefore, AGM can be considered as a significant progress in nonlinear sciences. 展开更多
关键词 nonlinear differential equation Akbari-Ganji's method(AGM) rigid beam viscoelastic foundation vibrating system damping ratio energy lost per cycle
下载PDF
Scattering loss in optical waveguide with trapezoidal cross section 被引量:1
17
作者 彭文强 吴宇列 +1 位作者 刘勇 尹自强 《Journal of Central South University》 SCIE EI CAS 2012年第5期1317-1321,共5页
To get the scattering loss of the trapezoidal core waveguide,a new analysis model is presented based on the perturbation equivalent method and modified effective-index method.Firstly,the trapezoidal core waveguide is ... To get the scattering loss of the trapezoidal core waveguide,a new analysis model is presented based on the perturbation equivalent method and modified effective-index method.Firstly,the trapezoidal core waveguide is successfully equivalent to the rectangular one with both restricting the same optical field energy by adopting the perturbation method,Then,the equivalent rectangular core waveguide is decomposed into two slab waveguides by employing the modified effective-index method,The trapezoidal core waveguide scattering theory model is established based on the slab waveguide scattering theory.With the sidewalls surface roughness in the range from 0 to 100 nm in the single model trapezodial core waveguide,optical simulation shows excellent agreement with the results from the scattering loss model presented.The relationship between the dimension and side-wall roughness with the scattering loss can be determined in the trapezoidal core waveguide by the scattering loss model. 展开更多
关键词 trapezoidal core waveguide perturbation equivalent method modified effective-index method scattering loss
下载PDF
径向基函数神经网络分类器与CNN在癌细胞图像分类中的应用 被引量:1
18
作者 张凯凯 郭松林 毕晨琳 《电子测试》 2019年第22期66-67,76,共3页
本文提出一种新的用于对卷积神经网络提取的特征进行分类的分类器即径向基函数神经网络(rbfnn)分类器。其思想是利用卷积神经网络作为特征提取器,使用rbfnn对提取的特征进行分类。同时在训练时采取softmax分类器与rbfnn分类器同步训练... 本文提出一种新的用于对卷积神经网络提取的特征进行分类的分类器即径向基函数神经网络(rbfnn)分类器。其思想是利用卷积神经网络作为特征提取器,使用rbfnn对提取的特征进行分类。同时在训练时采取softmax分类器与rbfnn分类器同步训练的方式,其中rbfnn分类器将MSE(均方误差)损失作为监督信息,softmax分类器用交叉熵损失作为监督信息。优化后的模型优于[1]中的72.9%的准确率。 展开更多
关键词 卷积神经网络 径向基神经网络 图像识别 MSE(均误差)损失
下载PDF
Productivity prediction model and optimal configuration of herringbone multilateral well 被引量:1
19
作者 蔡明 尹洪军 +1 位作者 钟会影 付春权 《Journal of Central South University》 SCIE EI CAS 2013年第3期770-775,共6页
In order to accurately predict the productivity of herringbone multilateral well,a new productivity prediction model was founded.And based on this model,orthogonal test and multiple factor variance analysis were appli... In order to accurately predict the productivity of herringbone multilateral well,a new productivity prediction model was founded.And based on this model,orthogonal test and multiple factor variance analysis were applied to study optimization design of herringbone multilateral well.According to the characteristics of herringbone multilateral well,by using pressure superposition and mirror image reflection theory,the coupled model of herringbone multilateral well was developed on the basis of a three-dimensional pseudo-pressure distribution model for horizontal wells.The model was formulated in consideration of friction loss,acceleration loss of the wellbore and mixed loss at the confluence of main wellbore and branched one.After mathematical simulation on productivity of the herringbone multilateral well with the coupled model,the effects of well configuration on productivity were analyzed.The results show that lateral number is the most important factor,length of main wellbore and length of branched wellbore are the secondary ones,angle between main and branched one has the least influence. 展开更多
关键词 herringbone multilateral well wellbore pressure drop productivity prediction optimization design
下载PDF
A general solution and approximation for the diffusion of gas in a spherical coal sample 被引量:4
20
作者 Wang Yucang Xue Sheng Xie Jun 《International Journal of Mining Science and Technology》 SCIE EI 2014年第3期345-348,共4页
The square root relationship of gas release in the early stage of desorption is widely used to provide a simple and fast estimation of the lost gas in coal mines. However, questions arise as to how the relationship wa... The square root relationship of gas release in the early stage of desorption is widely used to provide a simple and fast estimation of the lost gas in coal mines. However, questions arise as to how the relationship was theoretically derived, what are the assumptions and applicable conditions and how large the error will be. In this paper, the analytical solutions of gas concentration and fractional gas loss for the diffusion of gas in a spherical coal sample were given with detailed mathematical derivations based on the diffusion equation. The analytical solutions were approximated in case of small values of time and the error analyses associated with the approximation were also undertaken. The results indicate that the square root relationship of gas release is the first term of the approximation, and care must be taken in using the square root relationship as a significant error might be introduced with increase in the lost time and decrease in effective diameter of a spherical coal sample. 展开更多
关键词 Coal content Lost gas Spherical coal sample Gas diffusion APPROXIMATION Error analysis
下载PDF
上一页 1 2 下一页 到第
使用帮助 返回顶部