本文基于因果推断理论,提出根据病人的生物标记物进行最优治疗方案选择的统计方法.这种方法是基于CATE(conditional average treatment effect)曲线以及CATE曲线的置信带(SCB)的.CSTE曲线表示给定生物标记物(协变量)的条件下,处理组的...本文基于因果推断理论,提出根据病人的生物标记物进行最优治疗方案选择的统计方法.这种方法是基于CATE(conditional average treatment effect)曲线以及CATE曲线的置信带(SCB)的.CSTE曲线表示给定生物标记物(协变量)的条件下,处理组的条件平均处理效应.同时,CATE曲线及其SCB可以被用于对特定的治疗方案选择适宜的病人.文中利用B样条方法估计CATE曲线及其CSB,并推导了其近似大样本性质.文中还通过模拟比较研究了CATE曲线的置信带的有限样本性质,并阐述了CATE曲线及其置信带在真实数据中如何选择最优治疗方案.展开更多
活化腐蚀产物是聚变堆的主要源项之一,由于聚变堆的中子通量密度是按短时脉冲产生的,使得进行长周期的源项分析时需要计算大量的脉冲,因此需要研究脉冲的等效处理方法以提高计算效率。目前可用的脉冲等效处理方法主要有稳态(SS)方法、...活化腐蚀产物是聚变堆的主要源项之一,由于聚变堆的中子通量密度是按短时脉冲产生的,使得进行长周期的源项分析时需要计算大量的脉冲,因此需要研究脉冲的等效处理方法以提高计算效率。目前可用的脉冲等效处理方法主要有稳态(SS)方法、等效稳态(ESS)方法和连续脉冲(CP)方法。本文在活化腐蚀产物源项分析程序CATE3.0的基础上,增加了脉冲等效处理功能,并对中国聚变工程实验堆(China Fusion Engineering Test Reactor, CFETR)的包层水冷回路进行了建模仿真。分别采用SS、ESS和CP三种方法对CFETR活化腐蚀产物源项进行计算,并以精确脉冲建模(EP)方法的计算结果为基准进行比较和分析。结果表明:在单级脉冲运行下,前期使用ESS方法近似处理的CP方法(CP(ESS))计算精度最高,半衰期为102 s~104 s的核素,当末尾的脉冲时间在10~15个左右半衰期后就可以趋于稳定;在多级脉冲一级等效下,CP(ESS)方法同样适用,并且末尾保留少量脉冲时,计算结果就能达到较高的精度;此外,脉冲时间越短,脉冲等效处理方法的计算时间节省效益越显著。Activated corrosion products are one of the main source terms of fusion reactor. Because the neutron flux density of fusion reactor is generated by short-time pulse, a large number of pulses are involved in the long-period source term analysis. So it is necessary to study the equivalent treatment method of pulses to improve the calculation efficiency. At present, the available pulse equivalent treatment methods mainly include steady state (SS) method, equivalent steady state (ESS) method and continuous pulse (CP) method. In this paper, on the basis of the activated corrosion products source term analysis program CATE3.0, the pulse equivalent treatment function was added, and simulated the blanket water cooling loop of the China Fusion Engineering Test Reactor (CFETR). SS, ESS and CP methods were used to calculate the source terms of CFETR activated corrosion products, and the calculation results of exact pulse (EP) modeling method were used as the basis for systematic comparison and analysis. The results show that in the single-level pulse operation, the CP method, which is approximated by ESS method in the early stage (CP(ESS)), has higher calculation accuracy. Nuclides with a half-life period of 102 s~104 s can stabilize when the final pulse time is about 10~15 half-life period. In the multistage pulse first-level equivalence, CP(ESS) method is also applicable, and it is highly accurate when a small number of pulses are retained at the end. In addition, the shorter the pulse time, the more significant the computational time saving benefit of pulse equivalent treatment methods.展开更多
Pt-based magnetic nano catalysts are one of the most suitable cand idates for electrocatalytic materials due to their high electrochemistry activity and retrievability.Unfortunately,the inferior durability prevents th...Pt-based magnetic nano catalysts are one of the most suitable cand idates for electrocatalytic materials due to their high electrochemistry activity and retrievability.Unfortunately,the inferior durability prevents them from being scaled-up,limiting their commercial applications.Herein,an antiferromagnetic element Mn was introduced into PtCo nanostructured alloy to synthesize uniform Mn-PtCo truncated octahedral nanoparticles(TONPs)by one-pot method.Our results show that Mn can tune the blocking temperature of Mn-PtCo TONPs due to its an tiferromag netism.At low temperatures,Mn-PtCo TONPs are ferromag netic,and the coercivity in creases gradually with in creasi ng Mn contents.At room temperature,the Mn-PtCo TONPs display superparamag netic behavior,which is greatly helpful for in dustrial recycling.Mn doping can not only modify the electronic structure of PtCo TONPs but also enhanee electrocatalytic performance for methanol oxidation reaction.The maximum specific activity of Mn-PtCo-3 reaches 8.1 A`m^-2,3.6 times of commercial Pt/C(2.2 A·m^-2)and 1.4 times of PtCo TONPs(5.6 A`m^-2),respectively.The mass activity decreases by only 30%after 2,000 cycles,while it is 45%and 99%(nearly inactive)for PtCo TONPs and commercial Pt/C catalysts,respectively.展开更多
基金supported by the National Natural Science Foundataion of China(Grant No.81773546)。
文摘本文基于因果推断理论,提出根据病人的生物标记物进行最优治疗方案选择的统计方法.这种方法是基于CATE(conditional average treatment effect)曲线以及CATE曲线的置信带(SCB)的.CSTE曲线表示给定生物标记物(协变量)的条件下,处理组的条件平均处理效应.同时,CATE曲线及其SCB可以被用于对特定的治疗方案选择适宜的病人.文中利用B样条方法估计CATE曲线及其CSB,并推导了其近似大样本性质.文中还通过模拟比较研究了CATE曲线的置信带的有限样本性质,并阐述了CATE曲线及其置信带在真实数据中如何选择最优治疗方案.
文摘活化腐蚀产物是聚变堆的主要源项之一,由于聚变堆的中子通量密度是按短时脉冲产生的,使得进行长周期的源项分析时需要计算大量的脉冲,因此需要研究脉冲的等效处理方法以提高计算效率。目前可用的脉冲等效处理方法主要有稳态(SS)方法、等效稳态(ESS)方法和连续脉冲(CP)方法。本文在活化腐蚀产物源项分析程序CATE3.0的基础上,增加了脉冲等效处理功能,并对中国聚变工程实验堆(China Fusion Engineering Test Reactor, CFETR)的包层水冷回路进行了建模仿真。分别采用SS、ESS和CP三种方法对CFETR活化腐蚀产物源项进行计算,并以精确脉冲建模(EP)方法的计算结果为基准进行比较和分析。结果表明:在单级脉冲运行下,前期使用ESS方法近似处理的CP方法(CP(ESS))计算精度最高,半衰期为102 s~104 s的核素,当末尾的脉冲时间在10~15个左右半衰期后就可以趋于稳定;在多级脉冲一级等效下,CP(ESS)方法同样适用,并且末尾保留少量脉冲时,计算结果就能达到较高的精度;此外,脉冲时间越短,脉冲等效处理方法的计算时间节省效益越显著。Activated corrosion products are one of the main source terms of fusion reactor. Because the neutron flux density of fusion reactor is generated by short-time pulse, a large number of pulses are involved in the long-period source term analysis. So it is necessary to study the equivalent treatment method of pulses to improve the calculation efficiency. At present, the available pulse equivalent treatment methods mainly include steady state (SS) method, equivalent steady state (ESS) method and continuous pulse (CP) method. In this paper, on the basis of the activated corrosion products source term analysis program CATE3.0, the pulse equivalent treatment function was added, and simulated the blanket water cooling loop of the China Fusion Engineering Test Reactor (CFETR). SS, ESS and CP methods were used to calculate the source terms of CFETR activated corrosion products, and the calculation results of exact pulse (EP) modeling method were used as the basis for systematic comparison and analysis. The results show that in the single-level pulse operation, the CP method, which is approximated by ESS method in the early stage (CP(ESS)), has higher calculation accuracy. Nuclides with a half-life period of 102 s~104 s can stabilize when the final pulse time is about 10~15 half-life period. In the multistage pulse first-level equivalence, CP(ESS) method is also applicable, and it is highly accurate when a small number of pulses are retained at the end. In addition, the shorter the pulse time, the more significant the computational time saving benefit of pulse equivalent treatment methods.
基金The work was supported by the National Natural Science Foundation(Nos.51625101,51431009,5180118&and 51701202)the State Key Development Program for Basic Research of China(No.2015CB921401)+3 种基金the Fundamental Research Funds for the Central University Universities of China(No.FRF-TP-16-001C2)the China Postdoctoral Science Foundation(No.2018M632792)Startup Research Fund of Zhengzhou University(No.32210815)Bejing Natural Science Foundation(No.Z180014).
文摘Pt-based magnetic nano catalysts are one of the most suitable cand idates for electrocatalytic materials due to their high electrochemistry activity and retrievability.Unfortunately,the inferior durability prevents them from being scaled-up,limiting their commercial applications.Herein,an antiferromagnetic element Mn was introduced into PtCo nanostructured alloy to synthesize uniform Mn-PtCo truncated octahedral nanoparticles(TONPs)by one-pot method.Our results show that Mn can tune the blocking temperature of Mn-PtCo TONPs due to its an tiferromag netism.At low temperatures,Mn-PtCo TONPs are ferromag netic,and the coercivity in creases gradually with in creasi ng Mn contents.At room temperature,the Mn-PtCo TONPs display superparamag netic behavior,which is greatly helpful for in dustrial recycling.Mn doping can not only modify the electronic structure of PtCo TONPs but also enhanee electrocatalytic performance for methanol oxidation reaction.The maximum specific activity of Mn-PtCo-3 reaches 8.1 A`m^-2,3.6 times of commercial Pt/C(2.2 A·m^-2)and 1.4 times of PtCo TONPs(5.6 A`m^-2),respectively.The mass activity decreases by only 30%after 2,000 cycles,while it is 45%and 99%(nearly inactive)for PtCo TONPs and commercial Pt/C catalysts,respectively.
