刮板输送机直线度光纤精准感知与调直关键技术  

Key technologies of optical fiber accurate perception and straightening of straightness of the scraper

作　　者：方新秋[1] 陈宁宁 冯豪天 宋扬 梁敏富[1] 吴刚[1] 张璠 李家璇 李猛[1] 乔富康 吴洋 FANG Xinqiu;CHEN Ningning;FENG Haotian;SONG Yang;LIANG Minfu;WU Gang;ZHANG Fan;LI Jiaxuan;LI Meng;QIAO Fukang;WU Yang(School of Mines,Research Center of Intelligent Mining,China University of Mining and Technology,Xuzhou,Jiangsu 221116,China)

机构地区：[1]中国矿业大学矿业工程学院,智能化开采研究中心,江苏徐州221116

出　　处：《采矿与安全工程学报》2023年第5期1043-1056,共14页Journal of Mining & Safety Engineering

基　　金：国家自然科学基金项目(51874276,52004273,52104167);江苏省自然科学基金青年基金项目(BK20200639);中央高校建设世界一流大学(学科)和特色发展引导专项资金项目(2022ZZCX01K3);中央高校基本科研业务费专项资金项目(2020ZDPY0209);江苏省研究生科研与实践创新计划项目(KYCX23-2775);中国矿业大学研究生创新计划项目(2023WLKXJ016)。

摘　　要：煤矿智能化是实现煤炭工业高质量发展的必由之路,围绕煤矿智能化开采技术、刮板输送机直线度感知与调直技术、光纤形状传感技术现状及发展趋势,介绍了刮板输送机直线度光纤精准感知与调直的关键技术。针对刮板输送机形态感知不全面,感知信息可靠性差、感知精度低以及刮板输送机直线度调直不具备实时性、调直误差大等问题,基于“科学采矿”和“煤炭精准开采”理念,提出了智能工作面刮板输送机直线度光纤精准感知与调直的研究技术路线。首先建立刮板输送机形态监测模型并开发光纤形状传感重构算法,然后研究在纯弯曲、旋转以及扭转受力下光纤形状传感器中心波长的响应特性,提出光纤形状传感器纯弯曲与旋转、扭转分离的理论方法,构建光纤形状传感器的误差补偿模型;其次研究基于LSTM神经网络算法的刮板输送机直线度误差预测模型,对刮板输送机的目标轨迹进行动态预测,建立刮板输送机目标调直轨迹的修正模型和刮板输送机调直理论模型;最后开展了刮板输送机直线度光纤感知系统的具体工程应用,实现了刮板输送机形态的实时曲线显示、历史数据及曲线查询、数据存储、自动预警/报警等功能。研究成果为研究智能化工作面刮板输送机形态感知与调直控制提供理论基础和现场实践经验。Intelligent coal mine is the way to realize the advanced development of coal industry.Based on the coal mine intelligent mining technology,the scraper conveyor straightness monitoring and straightening technology,as well as the present and trend of optical fiber shape sensing technology,the key technologies of optical fiber accurate perception and straightening of straightness of scraper conveyor are introduced.To resolve the problems of incomplete shape perception of scraper conveyor,poor relia-bility and low accuracy of perception information,as well as the lack of real-time performance and large straightening errors of straightness of scraper conveyor,the research technical route of optical fiber accurate perception and straightening of straightness of scraper conveyor in intelligent working face is proposed based on the perception of scientific mining and precise coal mining.The shape monitoring model of scraper conveyor is established,and the reconstruction algorithm of optical fiber shape sensor and the response characteristics of the center wavelength of optical fiber shape sensor under pure bending,rotation and torsion are studied.The theoretical method of pure bending,rotation and torsion separation of optical fiber shape sensor is proposed,the error compensation model of optical fiber shape sensor is constructed,and model of scraper conveyor for the straightness error is established to predict the target trajectory of scraper conveyor dynamically based on LSTM neural network algorithm.The correction model of the target straightening track of the scraper conveyor and the theoretical model of the scraper conveyor straightening are established,and the specific engineering application of the scraper conveyor straightness fiber sensing system is carried out,which realizes real-time curve display,historical data and curve query,data storage and automatic warning/alarm of the scraper conveyor shape.It provides study of intelligent face scraper conveyor form perception and alignment control theoretically and pr

关 键 词：智能工作面 刮板输送机直线度 光纤光栅形状传感 误差补偿 调直方法 

分 类 号：TD325[矿业工程—矿井建设]