基于深度学习的液压支架支护质量评价方法研究  

Research on Hydraulic Support Supporting Quality Evaluation Method Based on Deep Learning

作　　者：王东亮 常亚军 陈文昱 Wang Dongliang;Chang Yajun;Chen Wenyu(CHN Energy Yulin Energy Co.,Ltd.,Yulin 719300,China;Zhengzhou Hengda Intelligent Control Technology Co.,Ltd.,Zhengzhou 450000,China)

机构地区：[1]国能榆林能源有限责任公司,陕西榆林719300 [2]郑州恒达智控科技股份有限公司,郑州450000

出　　处：《煤矿机械》2024年第4期196-200,共5页Coal Mine Machinery

基　　金：国家自然科学基金(52204168);河南省高校科技创新团队(22IRTSTHN005)。

摘　　要：为提高液压支架支护质量评价方法的操作效率和精度,提出一种基于深度学习模型的评价方法,将液压支架支护质量的评价问题转化为图像分类问题。结果表明,该评价方法在计算复杂度降低1/2的条件下取得了出色的分类精度(82.74%)和准确的定位精度(81.67%),具有明显的计算优势。In order to improve the operational efficiency and accuracy of the evaluation method for the supporting quality of hydraulic support,proposed an evaluation method based on the deep learning model,which the evaluation problem of hydraulic support supporting quality is transformed into an image classification problem.The results show that this evaluation method can achieve a classification accuracy of 89.07%and a positioning accuracy of 81.67%,which has obvious calculation advantage.

关 键 词：液压支架 支护质量 深度学习 评价方法 类激活映射 

分 类 号：TD355.4[矿业工程—矿井建设]