首页 | 官方网站   微博 | 高级检索  
     

基于立体视觉的数字图像相关方法在爆破抛掷作用研究中的应用
引用本文:崔新男,汪旭光,王尹军,张小军.基于立体视觉的数字图像相关方法在爆破抛掷作用研究中的应用[J].四川大学学报(工程科学版),2020,52(1):102-109.
作者姓名:崔新男  汪旭光  王尹军  张小军
作者单位:中国矿业大学(北京)
基金项目:中国工程院咨询研究项目 “智能爆破发展战略研究”(2018-XY-12)
摘    要:爆破抛掷物能否按照设计运动直接决定爆破工程的成败。为进一步研究爆破抛掷作用过程,量化研究抛掷物运动状态,本文采用数字图像相关和立体视觉原理,建立了爆破抛掷物运动观测系统。应用该系统观测了混凝土模型爆破试验过程,通过数字图像相关匹配算法进行抛掷物的跟踪,通过立体视觉技术计算抛掷物三维运动轨迹坐标,进而得到了其运动速度。结果表明,抛掷物的运动过程可分为整体加速运动和分散减速运动两个阶段。在第一阶段:在爆生气体推动下,抛掷物以一整体做加速运动,此阶段持续时间9ms,在6.5ms时抛掷物脱离爆破漏斗,最大速度可达16.63~19.65m/s;在第二阶段:抛掷物逐渐破裂、分散成块,处于不同区域、不同形状的碎块运动状态均不同,表面碎块的速度最大,可达26.24m/s,是潜在的爆破飞石,其只受到重力和空气阻力作用,做减速运动;中下部碎块在残余气体推动下,同样做减速运动,但减速缓慢,速度为9~16m/s;薄片状碎块在飞行中不断翻转,消耗自身动能,其中速度最小的仅为5~6m/s。整体运动阶段简单,而分散运动过程较复杂,除爆破参数、介质自身性质外,抛掷物碎块的形状、所处位置也会影响其运动状态,抛掷物表面碎块速度最大,更易形成爆破飞石。本文为量化研究爆破抛掷作用和预测爆破飞石提供了一种有效的新方法。

关 键 词:数字图像相关,抛掷作用,立体视觉,爆破飞石
收稿时间:2019/5/11 0:00:00
修稿时间:2019/10/25 0:00:00

Application of Digital Image Correlation Method Based on Stereo Vision in Blasting Throwing Effect
CUI Xinnan,WANG Xuguang,WANG Yinjun and ZHANG Xiaojun.Application of Digital Image Correlation Method Based on Stereo Vision in Blasting Throwing Effect[J].Journal of Sichuan University (Engineering Science Edition),2020,52(1):102-109.
Authors:CUI Xinnan  WANG Xuguang  WANG Yinjun and ZHANG Xiaojun
Affiliation:School of Mechanics and Civil Eng., China Univ. of Mining and Technol. (Beijing), Beijing 100083, China,School of Mechanics and Civil Eng., China Univ. of Mining and Technol. (Beijing), Beijing 100083, China;BGRIMM Technol. Group, Beijing 100160, China,BGRIMM Technol. Group, Beijing 100160, China and School of Civil and Resource Eng., Univ. of Science and Technol. Beijing, Beijing 100083, China
Abstract:Whether the blasting throwing object can move in accordance with design directly determines the success or failure of the blasting project. For further investigating the motion process of blasting throwing object, a new quantitative motion measurement system for blasting throwing effect based on digital image correlation and stereo vision was established in this paper. This system was used in several concrete model blasting tests for blasting throwing object tracking and observing. The throwing fragments were tracked by using digital image correlation matching algorithm and their special coordinates were calculated with stereo vison principle. Therefore, the trajectory and velocity of each fragment was obtained. The result shows that the motion process of blasting throwing object can be divided into two stages: the integral accelerating stage and dispersive decelerating stage. In the first stage, the thrust from explosion gas accelerates the integral throwing object with a maximum velocity of 16.63~19.65m/s. The termination of this stage is about 9ms and the throwing object detaches from the concrete model within 6.5ms. In the second stage, the throwing object falls into fragments which decelerate separately. The motion process of each fragment varies from each other because of its location and shape. Fragments on the surface are the potential blasting flyrocks as they have the largest velocity up to 26.24m/s. These fragments decelerate under gravity and air resistance. Fragments in the middle or lower place decelerate under gravity, air resistance and thrust from the residual explosion gas with a lower deceleration. And the velocity of these fragments is 9~16m/s. The flaky shaped fragments have the lowest velocity with 5~6m/s as they are spinning constantly during the motion process which consumes their kinetic energy. The integral motion is simple and the dispersive motion is complicated. In addition to the blasting parameters and characters of the medium itself, the shape and location of the throwing fragment will also influence its motion process. Fragments on the surface have the largest velocity and are easier to form blasting flyrocks. This paper provides an effective new method for blasting throwing effect research and flyrock prediction.
Keywords:digital image correlation  throwing effect  stereo vision  blasting flyrock
点击此处可从《四川大学学报(工程科学版)》浏览原始摘要信息
点击此处可从《四川大学学报(工程科学版)》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号