Pellet Size Detection and State Analysis based on Machine Vision and Intelligent Algorithm

Abstract

Aiming at the accuracy and efficiency of pellet size detection and state analysis in the process of iron and steel smelting, this paper proposes a solution based on machine vision and intelligent algorithm. The pellet images are collected by high-resolution camera equipment, and the particle size is automatically detected by convolutional neural network (CNN) model. At the same time, combining time series analysis and state space modeling, real-time monitoring and prediction algorithm is designed to quantify the relationship between pellet size and operation parameters. The experimental results show that the accuracy of particle size detection is more than 93%, and the production parameters can be adjusted in real time, which significantly improves the production efficiency and product quality. This study provides technical support for the intelligent upgrading of the iron and steel smelting industry, and lays a foundation for the optimization and expansion of the application of subsequent intelligent algorithms.