Sampling method for bulk coal car

Abstract

The invention discloses a bulk coal vehicle sampling method. The method comprises the following steps: step 1, preparing a contrast sample data set: 1.1, collecting sample data; 1.2, manually annotating portraits in the sample image library; 1.3, training a YOLO model by using the sample data set obtained in the step 1.2, and storing the updated YOLO sample contrast data set to a central control system; step 2, measuring the height of a platform floor; step 3, carrying out data acquisition; step 4, carrying out sampling point coordinate calculation; and step 5, collecting a sample. According to the bulk coal vehicle sampling method provided by the invention, a YOLO deep neural network learning mode is adopted, obstacles such as tie bars and compartment plates of a coal car compartment are intelligently identified, a sampling accident caused by collision between a sampling head and the obstacles is prevented, the height of a compartment bottom plate can be measured, the position of the lowest sampling point can be determined, and the sampling effect is ensured.

Description

technical field [0001] The invention relates to a method for sampling a loose coal car, belonging to the technical field of coal sampling. Background technique [0002] In coal-consuming enterprises such as thermal power plants and steel mills, the sampling of bulk coal entering the plant plays a decisive role in ensuring the quality of the production plant. At present, bulk coal is generally transported into the factory by bulk coal trucks, and the sampling needs to be carried out on the coal truck. Manual sampling is a heavy workload, and it is impossible to get coal samples in the middle and bottom of the truck. The samples taken are not representative. powerful. Then some mechanical sampling machines appeared. The commonly used automobile coal sampling device is a semi-closed structure, which is arranged in parallel above the compartment to be collected. The sampling head of the sampling machine is set on the sliding cable, and the sampling point is manually controlled....

AI Technical Summary

Problems solved by technology

However, the current coal transport vehicles have a large transportation volume, so there are often multiple reinforcement bars in the compartment to reinforce the side panels. If there is negligence and improper labeling, accidents of collision between the sampling head and the tie tend to occur, resulting in sampling failure or even equipment damage

And in order to prevent the sampling head from colliding with the floor of the carriage, the sampling machine needs to input the data of the height of the floor of the carriage before sampling. Since the height of the coal truck is different, the height of the floor of the carriage is also different. Therefore, it is necessary to measure the height of the bottom of the carriage for each vehicle. Clear height, the current measurement method is to use manual measurement and manually input the data into the coal sampling machine. This measurement method requires frequent operations by the operator and cannot be unattended, and manual operation creates opportunities for sampling cheating.

Moreover, due to problems such as uneven tire pressure and shock absorption and deformation of the carriage, the carriage is often inclined after loading coal. For the sampling of inclined carriages, some coal seams are located below the measurement height, so only one carriage height is currently used. It is difficult to achieve the purpose of sampling all faults in the coal body

Images