In October 2022,machine learning experts at Deep Mind(London,UK),a subsidiary of Google(Mountain View,CA,USA),reported a“breakthrough”on an extremely common mathematical algorithm called matrix multiplication[1].In ...In October 2022,machine learning experts at Deep Mind(London,UK),a subsidiary of Google(Mountain View,CA,USA),reported a“breakthrough”on an extremely common mathematical algorithm called matrix multiplication[1].In previous years,DeepMind has made headlines with its successes in using deep learning to master various games,such as Go[2],chess,and even the strategic board game Diplomacy,and,in a more recent and clearly practical application with its Alpha Fold program[3].展开更多
As the risks associated with air turbulence are intensified by climate change and the growth of the aviation industry,it has become imperative to monitor and mitigate these threats to ensure civil aviation safety.The ...As the risks associated with air turbulence are intensified by climate change and the growth of the aviation industry,it has become imperative to monitor and mitigate these threats to ensure civil aviation safety.The eddy dissipation rate(EDR)has been established as the standard metric for quantifying turbulence in civil aviation.This study aims to explore a universally applicable symbolic classification approach based on genetic programming to detect turbulence anomalies using quick access recorder(QAR)data.The detection of atmospheric turbulence is approached as an anomaly detection problem.Comparative evaluations demonstrate that this approach performs on par with direct EDR calculation methods in identifying turbulence events.Moreover,comparisons with alternative machine learning techniques indicate that the proposed technique is the optimal methodology currently available.In summary,the use of symbolic classification via genetic programming enables accurate turbulence detection from QAR data,comparable to that with established EDR approaches and surpassing that achieved with machine learning algorithms.This finding highlights the potential of integrating symbolic classifiers into turbulence monitoring systems to enhance civil aviation safety amidst rising environmental and operational hazards.展开更多
We establish a simulation model based on the theory of air flow to analyze the accelerated release effect of the quick release valve inside the air brake control valve.In addition, the combined simulation system of tr...We establish a simulation model based on the theory of air flow to analyze the accelerated release effect of the quick release valve inside the air brake control valve.In addition, the combined simulation system of train air brake system and longitudinal train dynamics is used to analyze how the parameters of the quick release valve in the 120/120–1 brake control valve affect the propagation characteristics of the train brake pipe pressure wave, the release action range of the accelerated brake, and the longitudinal coupler force for a 20,000-ton heavy haul train on the section of the Datong–Qinhuangdao Railway. The results show that the quick release valve can effectively accelerate the rising speed of the train brake pipe pressure during the initial release, as the accelerated release effect is evident before the train brake pipe pressure reaches582 k Pa. The quick release valve can effectively accelerate the release of the rear cars, reducing the longitudinal coupler force impact due to time delay of the release process. The quick release valve can effectively reduce the tensile coupler force in the train by as much as 20% in certain cases.展开更多
The Lijiang 2.4 m Telescope(LJT)is one of the most important telescopes for general astronomical observations in China.Yunnan Faint Object Spectrograph and Camera(YFOSC)is a widely used instrument mounted on the LJT,w...The Lijiang 2.4 m Telescope(LJT)is one of the most important telescopes for general astronomical observations in China.Yunnan Faint Object Spectrograph and Camera(YFOSC)is a widely used instrument mounted on the LJT,which occupies for~80%of the observing time of the LJT,and thus instrument health and data quality of the YFOSC are very important for both the telescope maintenance team and users of the LJT.So we develop an automated data analysis system for the quality control(QC)of YFOSC data.This system is also a new function of the observing support of the YFOSC.Based on the system,YFOSC data can be reduced quickly as they are acquired and QC parameters are extracted.Observers can assess the quality of their data and make a possible revision of their observing plan in time.These parameters can also be used to check the health of the YFOSC,which is helpful for the telescope maintenance team to find potential problems.All of these aim at improving the productivity of the LJT.展开更多
Objective:Fournier’s gangrene is a rare but life-threatening infection disease with high mortality rate.The quick Sepsis-related Organ Failure Assessment(qSOFA)is a new and simpler scoring system that may identify pa...Objective:Fournier’s gangrene is a rare but life-threatening infection disease with high mortality rate.The quick Sepsis-related Organ Failure Assessment(qSOFA)is a new and simpler scoring system that may identify patients with suspected infection who are at greater risk for a poor outcome.The purpose of this study was to find out role of qSOFA in determining prognosis of Fournier’s gangrene patients.Methods:This study is a case control with retrospective review of Fournier’s gangrene patients treated at Hasan Sadikin Hospital from January 2013 to December 2017 who met inclusion criteria.Participants were divided into two groups according to qSOFA score as high qSOFA(2-3)and low qSOFA(0-1).Results:From 69 patients,the mortality rate was 24.6%.The sensitivity of qSOFA score to predict mortality was 88.2%;the specificity was 94.2%;positive predictive value was 83.3%;negative predictive value was 96.1%;positive likelihood ratio was 15.2;negative likelihood ratio was 0.12;and the area under the receiver operating characteristic curve of qSOFA was 94.2%.There was significant association between qSOFA scale and mortality with p-value of 0.0001.The qSOFA score has strong positive correlation with Fournier’s Gangrene Severity Index(p<0.0001,r=0.704).Conclusion:qSOFA scoring system has a high prognostic value and can be used to determine prognosis of Fournier’s gangrene patients.展开更多
The scheme of robot positioning based on multiple quick response(QR)code landmarks is proposed.Firstly,the pose of the robot relative to the QR code landmarks is obtained by extracting the feature points of the QR cod...The scheme of robot positioning based on multiple quick response(QR)code landmarks is proposed.Firstly,the pose of the robot relative to the QR code landmarks is obtained by extracting the feature points of the QR code,combined with the visual computing technology.Then,the conversion of the absolute pose of the robot is completed based on the absolute position of the QR code landmark,and thus the real-time position of the robot is obtained.The proposed scheme makes full use of the QR code fault tolerance and multiple QR code landmarks to improve the calculation accuracy.It has good robustness and versatility,and it is easy to implement.It can help the robot to complete the positioning in the actual work,making robot navigation more accurate.展开更多
概念漂移是流数据挖掘领域中的一个重要且具有挑战性的难题.然而,目前的方法大多仅能够处理线性或简单的非线性映射,深度神经网络虽然有较强的非线性拟合能力,但在流数据挖掘任务中,每次只能在新得到的1个或一批样本上进行训练,学习模...概念漂移是流数据挖掘领域中的一个重要且具有挑战性的难题.然而,目前的方法大多仅能够处理线性或简单的非线性映射,深度神经网络虽然有较强的非线性拟合能力,但在流数据挖掘任务中,每次只能在新得到的1个或一批样本上进行训练,学习模型难以实时调整以适应动态变化的数据流.为解决上述问题,将梯度提升算法的纠错思想引入含概念漂移的流数据挖掘任务之中,提出了一种基于自适应深度集成网络的概念漂移收敛方法(concept drift convergence method based on adaptive deep ensemble networks,CD_ADEN).该模型集成多个浅层神经网络作为基学习器,后序基学习器在前序基学习器输出的基础上不断纠错,具有较高的实时泛化性能.此外,由于浅层神经网络有较快的收敛速度,因此所提出的模型能够较快地从概念漂移造成的精度下降中恢复.多个数据集上的实验结果表明,所提出的CD_ADEN方法平均实时精度有明显提高,相较于对比方法,平均实时精度有1%~5%的提升,且平均序值在7种典型的对比算法中排名第一.说明所提出的方法能够对前序输出进行纠错,且学习模型能够快速地从概念漂移造成的精度下降中恢复,提升了在线学习模型的实时泛化性能.展开更多
文摘In October 2022,machine learning experts at Deep Mind(London,UK),a subsidiary of Google(Mountain View,CA,USA),reported a“breakthrough”on an extremely common mathematical algorithm called matrix multiplication[1].In previous years,DeepMind has made headlines with its successes in using deep learning to master various games,such as Go[2],chess,and even the strategic board game Diplomacy,and,in a more recent and clearly practical application with its Alpha Fold program[3].
基金supported by the Meteorological Soft Science Project(Grant No.2023ZZXM29)the Natural Science Fund Project of Tianjin,China(Grant No.21JCYBJC00740)the Key Research and Development-Social Development Program of Jiangsu Province,China(Grant No.BE2021685).
文摘As the risks associated with air turbulence are intensified by climate change and the growth of the aviation industry,it has become imperative to monitor and mitigate these threats to ensure civil aviation safety.The eddy dissipation rate(EDR)has been established as the standard metric for quantifying turbulence in civil aviation.This study aims to explore a universally applicable symbolic classification approach based on genetic programming to detect turbulence anomalies using quick access recorder(QAR)data.The detection of atmospheric turbulence is approached as an anomaly detection problem.Comparative evaluations demonstrate that this approach performs on par with direct EDR calculation methods in identifying turbulence events.Moreover,comparisons with alternative machine learning techniques indicate that the proposed technique is the optimal methodology currently available.In summary,the use of symbolic classification via genetic programming enables accurate turbulence detection from QAR data,comparable to that with established EDR approaches and surpassing that achieved with machine learning algorithms.This finding highlights the potential of integrating symbolic classifiers into turbulence monitoring systems to enhance civil aviation safety amidst rising environmental and operational hazards.
基金China National Railway Group Co.,Ltd(N2020J037).
文摘We establish a simulation model based on the theory of air flow to analyze the accelerated release effect of the quick release valve inside the air brake control valve.In addition, the combined simulation system of train air brake system and longitudinal train dynamics is used to analyze how the parameters of the quick release valve in the 120/120–1 brake control valve affect the propagation characteristics of the train brake pipe pressure wave, the release action range of the accelerated brake, and the longitudinal coupler force for a 20,000-ton heavy haul train on the section of the Datong–Qinhuangdao Railway. The results show that the quick release valve can effectively accelerate the rising speed of the train brake pipe pressure during the initial release, as the accelerated release effect is evident before the train brake pipe pressure reaches582 k Pa. The quick release valve can effectively accelerate the release of the rear cars, reducing the longitudinal coupler force impact due to time delay of the release process. The quick release valve can effectively reduce the tensile coupler force in the train by as much as 20% in certain cases.
基金supported by the National Natural Science Foundation of China(U1931131)West Light Foundation of Chinese Academy of Sciences。
文摘The Lijiang 2.4 m Telescope(LJT)is one of the most important telescopes for general astronomical observations in China.Yunnan Faint Object Spectrograph and Camera(YFOSC)is a widely used instrument mounted on the LJT,which occupies for~80%of the observing time of the LJT,and thus instrument health and data quality of the YFOSC are very important for both the telescope maintenance team and users of the LJT.So we develop an automated data analysis system for the quality control(QC)of YFOSC data.This system is also a new function of the observing support of the YFOSC.Based on the system,YFOSC data can be reduced quickly as they are acquired and QC parameters are extracted.Observers can assess the quality of their data and make a possible revision of their observing plan in time.These parameters can also be used to check the health of the YFOSC,which is helpful for the telescope maintenance team to find potential problems.All of these aim at improving the productivity of the LJT.
文摘Objective:Fournier’s gangrene is a rare but life-threatening infection disease with high mortality rate.The quick Sepsis-related Organ Failure Assessment(qSOFA)is a new and simpler scoring system that may identify patients with suspected infection who are at greater risk for a poor outcome.The purpose of this study was to find out role of qSOFA in determining prognosis of Fournier’s gangrene patients.Methods:This study is a case control with retrospective review of Fournier’s gangrene patients treated at Hasan Sadikin Hospital from January 2013 to December 2017 who met inclusion criteria.Participants were divided into two groups according to qSOFA score as high qSOFA(2-3)and low qSOFA(0-1).Results:From 69 patients,the mortality rate was 24.6%.The sensitivity of qSOFA score to predict mortality was 88.2%;the specificity was 94.2%;positive predictive value was 83.3%;negative predictive value was 96.1%;positive likelihood ratio was 15.2;negative likelihood ratio was 0.12;and the area under the receiver operating characteristic curve of qSOFA was 94.2%.There was significant association between qSOFA scale and mortality with p-value of 0.0001.The qSOFA score has strong positive correlation with Fournier’s Gangrene Severity Index(p<0.0001,r=0.704).Conclusion:qSOFA scoring system has a high prognostic value and can be used to determine prognosis of Fournier’s gangrene patients.
基金Project of Introducing Urgently Needed Talents in Key Support Areas of Shandong Province,China。
文摘The scheme of robot positioning based on multiple quick response(QR)code landmarks is proposed.Firstly,the pose of the robot relative to the QR code landmarks is obtained by extracting the feature points of the QR code,combined with the visual computing technology.Then,the conversion of the absolute pose of the robot is completed based on the absolute position of the QR code landmark,and thus the real-time position of the robot is obtained.The proposed scheme makes full use of the QR code fault tolerance and multiple QR code landmarks to improve the calculation accuracy.It has good robustness and versatility,and it is easy to implement.It can help the robot to complete the positioning in the actual work,making robot navigation more accurate.
文摘概念漂移是流数据挖掘领域中的一个重要且具有挑战性的难题.然而,目前的方法大多仅能够处理线性或简单的非线性映射,深度神经网络虽然有较强的非线性拟合能力,但在流数据挖掘任务中,每次只能在新得到的1个或一批样本上进行训练,学习模型难以实时调整以适应动态变化的数据流.为解决上述问题,将梯度提升算法的纠错思想引入含概念漂移的流数据挖掘任务之中,提出了一种基于自适应深度集成网络的概念漂移收敛方法(concept drift convergence method based on adaptive deep ensemble networks,CD_ADEN).该模型集成多个浅层神经网络作为基学习器,后序基学习器在前序基学习器输出的基础上不断纠错,具有较高的实时泛化性能.此外,由于浅层神经网络有较快的收敛速度,因此所提出的模型能够较快地从概念漂移造成的精度下降中恢复.多个数据集上的实验结果表明,所提出的CD_ADEN方法平均实时精度有明显提高,相较于对比方法,平均实时精度有1%~5%的提升,且平均序值在7种典型的对比算法中排名第一.说明所提出的方法能够对前序输出进行纠错,且学习模型能够快速地从概念漂移造成的精度下降中恢复,提升了在线学习模型的实时泛化性能.
