Random vertical track irregularities are one of essential vibration sources in bridge, track structure and high-speed train systems. The common model of such irregularities is a stationary and ergodic Gaussian process...Random vertical track irregularities are one of essential vibration sources in bridge, track structure and high-speed train systems. The common model of such irregularities is a stationary and ergodic Gaussian process. The study presents the results of numerical dynamic analysis of advanced virtual models of composite BTT （bridge/ballasted track structure/high-speed train） systems. The analysis has been conducted for a series of types of single-span simply-supported railway composite （steel-concrete） bridges, with a symmetric platform, located on lines with ballasted track structure adapted for high-speed trains. The bridges are designed according to Polish bridge standards. A new methodology of numerical modeling and simulation of dynamic processes in BTT systems has been applied. The methodology takes into consideration viscoelastic suspensions of rail-vehicles, nonlinear Hertz wheel-rail contact stiffness and one-side wheel-rail contact, physically nonlinear elastic-damping properties of the track structure, random vertical track irregularities, approach slabs and other features. Computer algorithms of FE （finite element） modeling and simulation were programmed in Delphi. Both static and dynamic numerical investigations of the bridges forming the series of types have been carried out. It has been proved that in the case of common structural solutions of bridges and ballasted track structures, it is necessary to put certain limitations on operating speeds, macadam ballast and vertical track roughness.展开更多
针对当前大田环境条件下对害虫进行识别研究的不足,以南方蔬菜重大害虫为研究对象,探索了一种在大田环境下使用黄色诱捕板对蔬菜害虫进行监测计数的方法。在经典图像处理算法基础上,根据害虫监测目标的需要,提出了一种基于结构化随机森...针对当前大田环境条件下对害虫进行识别研究的不足,以南方蔬菜重大害虫为研究对象,探索了一种在大田环境下使用黄色诱捕板对蔬菜害虫进行监测计数的方法。在经典图像处理算法基础上,根据害虫监测目标的需要,提出了一种基于结构化随机森林的害虫图像分割算法和利用不规则结构的特征提取算法,进一步结合背景去除、干扰目标去除和检测模型计数子算法,集成设计了基于视觉感知的蔬菜害虫计数算法(Vegetable pest counting algorithm based on visual perception,VPCA-VP)。使用了现场环境下拍摄的图像进行实验与分析,共识别出蓟马9351只,烟粉虱202只,实蝇23只。经过与人工计数比对得出,本文基于视觉感知的蔬菜害虫计数算法的平均识别正确率为94.89%。其中,蔬菜害虫蓟马的识别正确率为93.19%,烟粉虱的识别正确率为91%,实蝇的识别正确率达到100%。算法达到了较好的测试性能,可以满足害虫快速计数需求,在农田害虫监测中有一定的应用前景。展开更多
文摘Random vertical track irregularities are one of essential vibration sources in bridge, track structure and high-speed train systems. The common model of such irregularities is a stationary and ergodic Gaussian process. The study presents the results of numerical dynamic analysis of advanced virtual models of composite BTT （bridge/ballasted track structure/high-speed train） systems. The analysis has been conducted for a series of types of single-span simply-supported railway composite （steel-concrete） bridges, with a symmetric platform, located on lines with ballasted track structure adapted for high-speed trains. The bridges are designed according to Polish bridge standards. A new methodology of numerical modeling and simulation of dynamic processes in BTT systems has been applied. The methodology takes into consideration viscoelastic suspensions of rail-vehicles, nonlinear Hertz wheel-rail contact stiffness and one-side wheel-rail contact, physically nonlinear elastic-damping properties of the track structure, random vertical track irregularities, approach slabs and other features. Computer algorithms of FE （finite element） modeling and simulation were programmed in Delphi. Both static and dynamic numerical investigations of the bridges forming the series of types have been carried out. It has been proved that in the case of common structural solutions of bridges and ballasted track structures, it is necessary to put certain limitations on operating speeds, macadam ballast and vertical track roughness.
基金Projects(U1934207, 52078487, 51778630) supported by the National Natural Science Foundation of ChinaProject(2019RS3009) supported by the Innovative Provincial Construction Project of Hunan,ChinaProject(2022ZZTS0150) supported by the Fundamental Research Funds of the Central Universities,China
文摘针对当前大田环境条件下对害虫进行识别研究的不足,以南方蔬菜重大害虫为研究对象,探索了一种在大田环境下使用黄色诱捕板对蔬菜害虫进行监测计数的方法。在经典图像处理算法基础上,根据害虫监测目标的需要,提出了一种基于结构化随机森林的害虫图像分割算法和利用不规则结构的特征提取算法,进一步结合背景去除、干扰目标去除和检测模型计数子算法,集成设计了基于视觉感知的蔬菜害虫计数算法(Vegetable pest counting algorithm based on visual perception,VPCA-VP)。使用了现场环境下拍摄的图像进行实验与分析,共识别出蓟马9351只,烟粉虱202只,实蝇23只。经过与人工计数比对得出,本文基于视觉感知的蔬菜害虫计数算法的平均识别正确率为94.89%。其中,蔬菜害虫蓟马的识别正确率为93.19%,烟粉虱的识别正确率为91%,实蝇的识别正确率达到100%。算法达到了较好的测试性能,可以满足害虫快速计数需求,在农田害虫监测中有一定的应用前景。
