Sand gradation detection method based on local sampling

Yang Zhang

doi:10.1038/s41598-024-80980-4

2024-11-27

DOI：10.1038/s41598-024-80980-4

原文链接

查看原文

摘要

En 中文

Sand gradation is an important reference factor for concrete proportion design, which has a significant impact on the strength and workability of concrete. The efficiency of using the traditional vibrating sieving method to measure the gradation is low, and the machine vision-based sand gradation inspection method can realize the automation of sand gradation inspection, which can significantly improve the inspection efficiency and reduce the inspection cost. The particle size of construction sand is usually between 0.075 and 4.75 mm, and to capture images of sand below 0.15 mm requires a high-resolution camera that can only maintain a small object distance and a small shooting field of view, so it is difficult for the image acquisition system to capture images of all the sand particles, and this is a challenging problem in sand grading inspection. In order to solve this problem, this study proposes a sand gradation detection method based on localized sampling, and develops a hardware system for localized sampling and a software system for gradation detection. The hardware system uses a flexible vibrating disk to uniformly disperse the sand particles, realizing random local sampling of sand particles, and calculating the grain size gradation of the test samples based on the local data of multiple local sampling. The experimental results show that the local sampling gradation and the sample gradation have a similar distribution, and the local gradation detection error of multiple sampling is smaller. This study provides an effective method for the automated detection of sand grain size gradation between 0.075 and 4.75 mm, which can significantly increase the frequency of construction sand gradation detection, improve the quality control level of construction sand, and improve the quality of construction.

Keyword:

Mechanized sand

Sand gradation detection

Local Sampling