论文部分内容阅读
针对煤样密度组成人工测量的滞后性,提出了一种基于图像分析的粗粒煤堆密度组成实时估计方法.引入煤堆图像定向分割算法和全局分割算法,提取了50个煤粒表面特征参数,根据其随密度级的变化趋势筛选出了3个有效特征参数,利用改进的KNN算法预测煤粒密度级,并结合煤粒质量模型实时估计煤堆密度组成.测试结果表明,粗粒煤堆密度组成估计的绝对误差最高为7.15%,最低为1.41%.
Aimed at the lag of artificial measurement of coal-like density composition, a real-time estimation method based on image analysis for coarse-grained coal bulk density was proposed. By introducing coal-pile image orientation segmentation algorithm and global segmentation algorithm, 50 coal particle surface feature parameters , Three effective characteristic parameters were selected according to the variation trend with the density grade, the coal density was predicted by the improved KNN algorithm, and the coal mass density composition was estimated in real time with the coal mass model.The test results showed that the coarse coal The absolute error of density component estimation is up to 7.15% and the lowest is 1.41%.