[1]王传奇,秦 伟,黄焕晟,等.基于HT-YOLO11的皮带跑偏智能识别及落料溜槽调节的主动防御策略[J].机械与电子,2026,44(08):35-42.
　WANG Chuanqi,QIN Wei,HUANG Huansheng,et al.Intelligent Recognition of Belt Deviation Based on HT-YOLO11 and Active Defense Strategy via Discharge Chute Adjustment[J].Machinery & Electronics,2026,44(08):35-42.
点击复制

基于HT-YOLO11的皮带跑偏智能识别及落料溜槽调节的主动防御策略()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
44
期数:
2026年08期
页码:
35-42
栏目:
智能检测
出版日期:
2026-08-25

文章信息/Info

Title:
Intelligent Recognition of Belt Deviation Based on HT-YOLO11 and Active Defense Strategy via Discharge Chute Adjustment
文章编号:
1001-2257(2026)08-0035-08
作者:
王传奇1; 秦 伟1; 黄焕晟1; 唐锐杰2; 万书亭2
1.国能珠海港务有限公司,广东 珠海 519040; 2.华北电力大学河北省电力机械装备健康维护与失效预防重点实验室,河北 保定 071003
Author(s):
WANG Chuanqi1; QIN Wei1; HUANG Huansheng1; TANG Ruijie2; WAN Shuting2
(1.Guoneng Zhuhai Port Co.,Ltd.,Zhuhai 519040,China;2.Hebei Key Laboratory of Electric Machinery Health Maintenance and Failure Prevention,North China Electric Power University,Baoding 071003,China)
关键词:
带式输送机; 皮带跑偏检测; 主动防御; HT YOLO11; 边缘线检测
Keywords:
belt conveyor; belt deviation detection; active defense strategy; HT-YOLO11; edge line detection
分类号:
TD714.4;TP391
文献标志码:
A
摘要:
针对传统皮带跑偏检测抗干扰性差、易受环境影响的缺陷,提出了基于托辊与皮带边缘交点坐标的检测方法,并基于YOLOv11构建改进网络。在骨干网络嵌入HEFS模块,通过多尺度池化、边缘增强与DSM 机制提取并筛选边缘特征;在检测头中采用TODTAH 解耦分类与回归任务,并结合动态可变形卷积提升细长边缘定位精度。实验结果表明,改进HT-YOLO11的mAP50、精确率与召回率均有显著提升,可精准定量识别皮带跑偏。进一步地,依据燃煤转运系统落料点偏移导致皮带跑偏的机理,提出了通过落料溜槽集料板调节落料点的皮带跑偏主动防御策略,与基于纠偏托辊的被动纠偏策略相比,可有效避免被动纠偏对皮带的损伤。
Abstract:
To overcome the limitations of traditional belt deviation detection methods,such as poor anti-interference capability and high susceptibility to environmental factors,this paper proposes a detection method based on the coordinates of intersection points between the idler and the belt edge,and constructs an improved network based on YOLOv11.A HEFS module is embedded into the backbone to extract and filter edge features through multi-scale pooling,edge enhancement and DSM mechanism.In the detection head, a TODTAH module is employed to decouple the classification and regression features,and the dynamic deformable convolution is incorporated to improve the localization accuracy of slender edges.Experimental results show that the improved HT-YOLO11 achieves significant improvements in the mAP50,precision, and recall,enabling accurate and quantitative identification of belt deviation.Furthermore,based on the mechanism by which the deviation of the material dropping point in the coal transfer systems induces belt deviation,an active defense strategy is proposed that adjusts the dropping point by means of the collection plate of discharge chute.Compared with the existing passive correction strategy based on self-aligning idlers,this active strategy can effectively avoid the belt damage caused by passive correction.

参考文献/References:

[1] 赵月爱,白渊铭,王玲,等.基于深度学习与计算机视觉的皮带跑偏检测研究[J].电子器件,2026,49(1):128-135. [2] 王浩宇,王希鹏.基于机器视觉的带式输送机胶带跑偏检测方法研究[J].中国煤炭,2024,50(增刊2):158-163. [3] 王成辰,王小丫,郭乃宏,等.传送带偏移检测技术研究及 FPGA实现[J].计算机测量与控制,2022,30(9):40-45. [4] 杨林顺,董志勇.基于图像处理的输送带跑偏故障在线检测技术研究[J].煤炭工程,2020,52(10):116-120. [5] 曾飞,陶玉衡,苏俊彬,等.融合ResNet18和Deconvolution的输送带横向跑偏检测方法[J].现代制造工程,2023(8):121-126. [6] 安龙辉,王满利,张长森.基于改进 RT-DETR 的井下输送带跑偏故障检测算法[J].工矿自动化,2025,51(3):54-62. [7] 王哲,傅哲,曹朋军,等.基于锚点定位和跨层修正的输送带跑偏识别算法[J].煤矿安全,2025,56(8):228-236. [8] 左明明,张习,杨子豪,等.基于深度学习的输送带跑偏状态智能监测方法[J].工矿自动化,2024,50(12):166-172. [9] 尹兆明,樊智敏.基于多体动力学特性的输送带跑偏研究[J].机械工程学报,2020,56(1):37-46. [10] 张顺.基于ADAMS带式输送机的跑偏研究[D].西安:西安科技大学,2019. [11] 曹高生,李志强,蒋刚.矿用带式输送机跑偏控制系统研究[J].机械与电子,2019,37(3):59-61. [12] 余浩,刘红梅,田俊杰,等.基于EDEM 的输送带动态跑偏仿真分析[J],机械与电子,2022,40(8):8-13. [13] 马乙杰,姚艳萍,韩刚,等.基于RecurDyn的带式输送机跑偏[J].机械设计,2025,42(9):151-160. [14] 余浩,田俊杰,姚庆,等.偏载工况下带式输送机输送带跑偏及纠偏研究[J]. 机电工程,2022,39(7):996-1003. [15] 曾飞,苏俊彬,宋杰杰,等.偏载工况下输送带横向跑偏检测系统设计[J].仪表技术与传感器,2022(8):69-73.

相似文献/References:

[1]曹高生,李志强,蒋刚.矿用带式输送机跑偏控制系统研究[J].机械与电子,2019,(03):59.
　CAO Gaosheng,LI Zhiqiang,JIANG Gang.Research on Deviation Control System of Mine Belt Conveyor[J].Machinery & Electronics,2019,(08):59.
[2]吕晓伟,王占飞.基于多特征融合的煤矿带式输送机驱动系统状态评估方法研究[J].机械与电子,2026,44(07):46.
　LYU Xiaowei,WANG Zhanfe.Research on a State Assessment Method for the Drive System of Coal Mine Belt Conveyors Based on Multi-feature Fusion[J].Machinery & Electronics,2026,44(08):46.
[3]孙 强,刘广毅,哈斯铁尔·艾列西,等.基于托辊裸露特征的输煤皮带跑偏检测方法[J].机械与电子,2026,44(04):47.
　SUN Qiang,LIU Guangyi,HASITIEER Ailiexi,et al.A Conveyor Belt Deviation Detection Method in Coal Handling Systems Based onthe Exposed Idler Rollers Feature[J].Machinery & Electronics,2026,44(08):47.

备注/Memo

备注/Memo:
收稿日期:2026-04-21 作者简介:王传奇 (1985-),男,辽宁抚顺人,高级工程师,研究方向为港口设备技术管理;秦 伟 (1990-),男,吉林洮南人,工程师,研究方向为港口设备技术管理;黄焕晟 (1995-),男,广东湛江人,工程师,研究方向为港口设备技术管理;唐锐杰 (2003-),男,河南驻马店人,硕士研究生,研究方向为电力设备状态监测与故障诊断;万书亭 (1970-),男,山西长治人,博士,教授,博士研究生导师,研究方向为电力设备状态监测与故障诊断,E-mail:52450809@ncepu.edu.cn。
更新日期/Last Update: 2026-08-28