基于RGB阈值划分算法的铜阳极炉精炼氧化终点预报模型  

A Prediction Model for the Endpoint of Oxidation in Copper Anodes Furnace Refining Based on the RGB Threshold Segmentation Algorithm

作　　者：刘大方[1] 王恩志 舒波 徐建新[3,4] 余彬 陈习堂 高荣 LIU Dafang;WANG Enzhi;SHU Bo;XU Jianxin;YU Bin;CHEN Xitang;GAO Rong(Yunnan Copper Co.,Ltd.,Kunming 650224,China;Chuxiong Dianzhong Nonferrous Metals Co.,Ltd.,Chuxiong 675000,Yunnan,China;State Key Laboratory of Complex Nonferrous Metal Resources Clean Utilization,Kunming University of Science and Technology,Kunming 650093,China;Faculty of Metallurgical and Energy Engineering,Kunming University of Science and Technology,Kunming 650093,China)

机构地区：[1]云南铜业股份有限公司,云南昆明650224 [2]楚雄滇中有色金属有限责任公司,云南楚雄675000 [3]昆明理工大学复杂有色金属资源清洁利用国家重点实验室,云南昆明650093 [4]昆明理工大学冶金与能源工程学院,云南昆明650093

出　　处：《昆明理工大学学报(自然科学版)》2025年第2期9-18,26,共11页Journal of Kunming University of Science and Technology(Natural Science)

基　　金：国家重点研发计划项目(2022YFC3902000);国家自然科学基金项目(52166004);云南省重大科技项目(202202AG050007,202202AG050002)。

摘　　要：本研究旨在开发一种基于RGB阈值划分算法的铜阳极炉精炼氧化终点预报模型.通过计算机视觉和RGB图像阈值处理技术,捕获和分析铜样品表面和断面的颜色、纹理变化信息,以预测氧化终点.研究结果表明:综合利用表面纹理预测模型、断面特征模型以及直方图相似度、余弦相似度和哈希相似度判别模型,可有效判断氧化终点的达到时间;当表面纹理模型呈现全白色且无黑色部分,断面特征模型氧化像素占比大于80%,直方图相似度模型自相似度达90%以上,余弦及哈希相似度模型余弦相似度接近1且哈希相似度较小时,可确定铜阳极炉精炼过程已达到氧化终点.该模型在预测准确性和稳定性方面取得了显著进展,为铜冶炼工业提供了可靠的预测工具,有助于降低能源消耗并提高生产效率.This study aims to develop a prediction model for the endpoint of oxidation in the refining of copper anode furnaces based on an RGB threshold segmentation algorithm.By employing computer vision and RGB image threshold processing techniques,the color and texture changes on the surface and cross-section of copper samples are captured and analyzed to predict the endpoint of oxidation.The results indicate that by integrating surface texture prediction models,cross-sectional feature models,and histogram similarity,cosine similarity,and hash similarity discrimination models,the endpoint of oxidation can be effectively determined.The findings show that when the surface texture model appears entirely white without any black parts,the cross-sectional feature model has an oxidation pixel ratio greater than 80%,the histogram similarity model reaches a self-similarity of over 90%,and the cosine and hash similarity models exhibit a cosine similarity close to 1 and a low hash similarity,it can be confirmed that the refining process of the copper anode furnace has reached the endpoint of oxidation.This model has made significant progress in terms of predictive accuracy and stability,providing a reliable forecasting tool for the copper smelting industry,which helps to reduce energy consumption and improve production efficiency.

关 键 词：RGB阈值划分 铜阳极炉 氧化终点 颜色 纹理 预测模型 

分 类 号：TF811[冶金工程—有色金属冶金] TP391.41[自动化与计算机技术—计算机应用技术]