传统的矿用设备维护一般是根据以往经验制定周期性巡检计划，维修方法依赖于经验丰富的技术人员现场勘查、检修。但随着矿山智能化程度和测控水平不断提高，对矿用设备有效检修提出了更高要求。采煤机作为综采工作面的核心设备，提高其检修效率尤为重要。因此，本文对采煤机液压系统的状态监测、故障预测以及预测性维护方案进行研究。

分析采煤机液压系统常见故障及特征，制定采煤机液压系统数字孪生体搭建方案，明确预测性维护系统机理与实现流程，提出基于液压状态信号的状态监测、故障预测方法，构建采煤机液压系统数字孪生体，实现预测性维护系统功能，保障采煤机安全稳定运行。

针对传统状态监测方法存在图表复杂难懂的应用难点，综合考虑采煤机液压系统常见故障囊括范围，研究基于状态监测模型的设备动态数据直观获取方法，以较小的数据通道和通用的开发引擎，搭建三维可视化设备状态监测平台，在虚拟空间自主选择观察目标，对应展示采煤机液压系统组成结构与状态参数。

针对故障预测过程中出现的数据集冗杂问题，研究基于灰色粗糙集的双向数据约简方法，建立一种适用于采煤机液压系统的优化人工神经网络预测模型，带入煤矿工作面实际数据进行网格训练，获得采煤机液压系统各关键部件预测故障率，并对比四种常见预测方法给出的预测结果准确度，验证了该方案适用于本系统。针对预测结果给出采煤机液压系统巡检、小修建议，在保证常态巡检的情况下，若将该系统应用于当前煤矿，可望在两年内减少43.6%巡检和27.1%停机小修。

最后，根据采煤机液压系统实际维修流程，制定了基于混合现实（Mixed Reality，MR）的采煤机液压系统预测性维护策略。采用Matlab-MySQL-Unity3D联合编程方式，将预测结果推送至故障对策库，进而驱动辅助维修流程部署至HoloLens眼镜，操作人员佩戴眼镜进行预测性维护操作，实现虚拟空间与现实设备的维修交互。该系统已针对状态监测、故障预测与预测性维护功能进行实验验证，结果表明各模块均实现预期功能。

The traditional maintenance of mining equipment is generally based on past experience to formulate a periodic inspection plan, and the maintenance method relies on on-site inspection and maintenance by experienced technicians. However, with the continuous improvement of mine intelligence and measurement and control level, higher requirements are put forward for effective maintenance of mining equipment. As the core equipment of the fully mechanized mining face, it is particularly important to improve the maintenance efficiency of the shearer. Therefore, this paper studies the condition monitoring, fault prediction and predictive maintenance scheme of the shearer hydraulic system. Analyze the common faults and characteristics of the hydraulic system of the shearer, formulate a construction plan for the digital twin of the hydraulic system of the shearer, clarify the mechanism and implementation process of the predictive maintenance system, propose a condition monitoring and fault prediction method based on hydraulic state signals, and finally build a predictive maintenance system. Maintain the system to ensure the safe and stable operation of the shearer.

Aiming at the difficulties in the application of complex and difficult charts in traditional condition monitoring methods, comprehensively consider the scope of common faults in the hydraulic system of shearers, and study the intuitive acquisition method of equipment dynamic data based on the condition monitoring model. , build a three-dimensional visual equipment state monitoring platform, independently select the observation target in the virtual space, and correspondingly display the composition structure and state parameters of the shearer hydraulic system.

Aiming at the problem of redundant and complex data sets in the process of fault prediction, a bidirectional data reduction method based on gray rough sets is studied, and an optimized artificial neural network prediction model suitable for the hydraulic system of the shearer is established. Grid training was used to obtain the predicted failure rate of each key component of the shearer hydraulic system, and the accuracy of the prediction results given by the four common prediction methods was compared to verify that the scheme is suitable for this system. According to the prediction results, the inspection and minor repair suggestions for the hydraulic system of the shearer are given. Under the condition of ensuring the normal inspection, if the system is applied to the current coal mine, it is expected to reduce 43.6% inspection and 27.1% downtime for minor repairs within two years.

Finally, according to the actual maintenance process of the shearer hydraulic system, a predictive maintenance strategy for the shearer hydraulic system based on Mixed Reality (MR) is formulated. Using the Matlab-MySQL-Unity3D joint programming method, the prediction results are pushed to the fault countermeasure library, and then the auxiliary maintenance process is driven to deploy to the HoloLens glasses. The system has been validated for condition monitoring, fault prediction and predictive maintenance functions, and the results show that each module performs as expected.

[1] 李森，王峰，刘帅，等. 综采工作面巡检机器人关键技术研究[J]. 煤炭科学技术,2020, 48(07): 218-225.

[2] 彭苏萍. 我国煤矿安全高效开采地质保障系统研究现状及展望[J]. 煤炭学报,2020, 45(07): 2331-2345.

[3] 王国法，刘峰，孟祥军，等. 煤矿智能化(初级阶段)研究与实践[J]. 煤炭科学技术，2019, 47(08): 1-36.

[4] 张旭辉，张雨萌，王妙云，等. 基于混合现实的矿用设备维修指导系统[J]. 工况自化，2019, 45(06): 27-31.

[5] 陶飞，刘蔚然，张萌，等. 数字孪生五维模型及十大领域应用[J]. 计算机集成制造系统，2019, 25(01): 1-18.

[6] 陈龙，王晓，杨健健，等. 平行矿山:从数字孪生到矿山智能[J]. 自动化学报，2021, 47(07): 1633-1645.

[7] TAO F, QING L, et al. New IT Driven Service-Oriented Smart Manufacturing: Framework and Characteristics[J]. IEEE Transaction on Systems, Man and Cybernetics: Systems, 2017.

[8] Milan V, Maria J, Alois Z, et al. Modular Fault Ascription and Corrective Maintenance Using a Digital Twin[J]. IFAC PapersOnLine, 2018, 51(11).

[9] TAO F, CHENG J, QI Q, et al. Digital Twin-deiven Product Design, Manufacturing and Service with Big Data[J]. The International Journal of Advanced Manufacturing Technlogy, 2017.

[10] TAO F, ZHANG M. Digital twin shop-floor: an new shop-floor floor pagradim towards smart manufacturing [J] IEEE Access, 2017, 5: 20418-20427.

[11] 陶飞，程颖，程江峰，等. 数字孪生车间信息物理融合理论与技术[J]. 计算机集成制造系统，2017, 23(08): 1603-1611.

[12] LIU X, TANG L, ZENG H, et al. Smart Design and Simulation of Aerospace Control System Based on Digital Twin[J]. Journal of System Simulation, 2019, 31(03): 377-384.

[13] Aivaliotis P, Georgoulias K, Ricatto R, et al. Predictive maintenance framework: Implementation of local and cloud processing for multi-stage prediction of CNC machines’ health, I-ESA Conference, Workshop: Predictive Maintenance in Industry 4.0: Methodologies, tools and interoperable applications, 2018 .

[14] Grieves M, Vickers J. Digital twin: Mitigating unpredictable, undesirable emergent behavior in complex systems[M] Transdisciplinary perspectives on complex systems. Springer, Cham, 2017: 85-113.

[15] 刘亚威. 数字线索助力美空军装备研制[J]. 测控技术，2018, 37(09): 1-4.

[16] 陈明，梁乃明. 智能制造之路:数字化工厂. 北京:机械工业出版社，2016. 10.

[17] Tuegel E. The Airframe Digital Twin: Some Challenges to Realization[C] Aiaa/asme/asce/ahs/asc Structures, Structural Dynamics & Materials Conference Aiaa/asme/ahs Adaptive Structures Conference Aiaa. 2013.

[18] Steven J, Henderson, Feiner S. Opportunistic controls: leveraging natural affordances as tangible user interfaces for augmented reality[C]. The Symposium on Virtual Reality Software & Technology, 2008: 211-218.

[19] Ríos J, Hernández J C, Oliva M, et al. Product Avatar as Digital Counterpart of a Physical Individual Product: Literature Review and Implications in an Aircraft[C]. ISPECE. 2015: 657-666.

[20] Rosen R, Wichert G V, Lo G, et al. About The Importance of Autonomy and Digital Twins for the Future of Manufacturing. [J] Ifac Papersonline, 2015, 48(3):567-572.

[21] Tao F, Zhang M. Digital Twin Shop-Floor: A New Shop-Floor Paradigm Towards Smart Manufacturing[J]. IEEE Access, 2017, 5(99): 20418-20427.

[22] Tao F, Cheng J, Qi Q, et al. Digital twin-driven product design, manufacturing and service with big data[J]. The International Journal of Advanced Manufacturing Technology, 2017, 26(03): 14-19.

[23] 陶飞，张萌，程江峰，等. 数字孪生车间—一种未来车间运行新模式[J]. 计算机集成制造系统，2017, 23(01): 1-9.

[24] 陶飞，马昕，胡天亮. 数字孪生标准体系[J]. 计算机集成制造系统，2019, 25(10): 2405-2418.

[25] 刘魁，王潘，刘婷. 数字孪生在航空发动机运行维护中的应用[J]. 航空动力，2019(04): 70-74.

[26] 武颖，姚丽亚，熊辉，等. 基于数字孪生技术的复杂产品装配过程质量管控方法[J]. 计算机集成制造系统，2019, 25(06): 1568-1575.

[27] 陶飞，刘蔚然，张萌，等. 数字孪生五维模型及十大领域应用[J]. 计算机集成制造系统，2019, 25(01): 5-22.

[28] TAO F, ZHANG M, CHENG J, et al. Digital twin workshop:a new paradigm for future workshop[J]. Computer Integrated Manufacturing Systems, 2017, 25(01): 8-13.

[29] 吴钱昊. 基于数字孪生车间的生产车间设备监控和设备故障预警方法研究[D]. 合肥工业大学, 2020.

[30] 刘义，刘晓冬，焦曼，等. 基于数字孪生的智能车间管控[J]. 制造业自动化，2020, 42(07): 148-152.

[31] 孙萌萌. 飞机总装配生产线数字孪生系统若干关键技术研究[D]. 浙江大学，2019.

[32] 董雷霆，周轩，赵福斌，等. 飞机结构数字孪生关键建模仿真技术[J]. 航空学报，2021, 42(03): 113-141.

[33] 张玉良，张佳朋，王小丹，等. 面向航天器在轨装配的数字孪生技术[J]. 导航与控制，2018, 17(03): 75-82.

[34] 陶飞，刘蔚然，刘检华，等. 数字孪生及其应用探索[J]. 计算机集成制造系统，2018, 24(01): 1-18.

[35] 方圆，刘江，吕瑞强，等. 基于数字孪生的设备加工过程监测技术研究[J]. 航空制造技术，2021, 64(04): 91-96.

[36] 苏新瑞，徐晓飞，卫诗嘉，等. 数字孪生技术关键应用及方法研究[J]. 中国仪器仪表，2019(07): 47-5.

[37] 丁华，杨亮亮，杨兆建，等. 数字孪生与深度学习融合驱动的采煤机健康状态预测[J]. 中国机械工程，2020, 31(07): 815-82.

[38] 吴淼，李瑞，王鹏江，等. 基于数字孪生的综掘巷道并行工艺技术初步研究[J]. 煤炭学报，2020, 45(S1): 506-513.

[39] 邹亚男，来孝楠，王鑫，等. 矿用电铲的形性一体化数字孪生方法及其验证[J]. 机电产品开发与创新，2021, 34(04): 20-22.

[40] 张旭辉，王妙云，张雨萌，等. 数据驱动下的工业设备虚拟仿真与远程操控技术研究[J]. 重型机械, 2018, 345(05): 20-23.

[41] 张旭辉，董润霖，马宏伟，等. 基于虚拟现实的煤矿救援机器人远程控制技术[J]. 煤炭科学技术，2017, 45(05): 52-57.

[42] 张旭辉,张雨萌,王岩等.数字孪生驱动的设备维修MR辅助指导技术[J/OL].计算机集成制造系统，2021, 27(8): 9.

[43] 杜涛. Predix：工业互联网的阶段性界碑[J]. 中国工业评论，2015, 000(010): 76-81.

[44] 杜郡. 西门子推出Mindsphere开放工业云[J]. 轻工机械，2016(3): 48.

[45] 黄海松，陈启鹏，李宜汀，等. 数字孪生技术在智能制造中的发展与应用研究综述[J]. 贵州大学学报：自然科学版，2020, 37(5): 8.

[46] 迟焕磊，袁智，曹琰，等. 基于数字孪生的智能化工作面三维监测技术研究[J]. 煤炭科学技术，2021, 49(10): 153-161.

[47] 谭飏，张宇，刘丽冰，等. 面向动力学特性监测的主轴系统数字孪生体[J]. 中国机械工程，2020, 31(18): 2231-2238+2246.

[48] 韩将星. 6G时代数字孪生在无线电监测站的应用研究[J]. 通信技术，2021, 54(02): 352-362.

[49] 李聪波，孙鑫，侯晓博，等. 数字孪生驱动的数控铣削刀具磨损在线监测方法[J]. 中国机械工程，2022, 33(1): 10.

[50] Sahal R, Alsamhi S, Brown K. Blockchain-Empowered Digital Twins Collaboration: Smart Transportation Use Case[J]. Machines, 2021, 9(9) : 10-33.

[51] 孟光，尤明懿. 基于状态监测的设备寿命预测与预防维护规划研究进展[J]. 振动与冲击，2011, 30(08): 1-11.

[52] 王锐，吴广宁，高国强，等. 高压断路器的预防性维护研究[J]. 城市建设理论研究:电子版，2012, 000(012): 1-4.

[53] Samad M, Mohsen G, Sara S, et al. Metrics development and modelling the mixed reality and digital twin adoption in the context of Industry 4.0[J]. Engineering Construction and Architectural Management, 2021: 5-10.

[54] 王亚强. 基于数字孪生的压力机润滑系统故障预测方法研究[D]. 山东大学，2021.

[55] 法鲁克巴斯特（Farooq Basit）. 面向纺纱信息物理系统的多层级预测建模与优化研究[D]. 东华大学，2021.

[56] Aivaliotis P, Georgoulias K. Chryssolouris G. The use of Digital Twin for predictive maintenance in manufacturing[J]. International Journal of Computer Integrated Manufacturing, 2019, 32(11) : 56-61.

[57] Hu L, Liu Z, Tan J. A VR simulation framework integrated with multisource CAE analysis data for mechanical equipment working process[J]. Computers in Industry, 2018, 97: 85-96.

[58] Cai Y, Starly B, Cohen P, et al. Sensor Data and Information Fusion to Construct Digital-twins Virtual Machine Tools for Cyber-physical Manufacturing[J]. Procedia Manufacturing, 2017, 10: 1031-1042.

[59] Farid K, Geraldo F. Digital twin modeling for predictive maintenance of gearboxes in floating offshore wind turbine drivetrains[J]. Forschung im Ingenieurwesen, 2021(prepublish).

[60] 王妍，张太华. 基于数字孪生的优化概率神经网络变压器故障诊断[J]. 组合机床与自动化加工技术，2020(11): 20-23.

[61] 李福兴，李璐爔，彭友. 基于数字孪生的船舶预测性维护[J]. 船舶工程，2020, 42(S1): 117-120+396.

[62] Xu Y, Sun Y, Liu X, et al. A Digital-Twin-Assisted Fault Diagnosis Using Deep Transfer Learning[J]. IEEE Access, 2019: 1-1.

[63] 石慧，曾建潮. 基于寿命预测的预防性维护维修策略[J]. 计算机集成制造系统，2014, 20(05): 1133-1140.

[64] 李维伟. 采煤机牵引部液压系统设计[J]. 机床与液压, 2015, 43(20): 95-97+87.

[65] 江伟，王璐. 采煤机液压回路系统设计[J]. 现代工业经济和信息化，2019, 9(09): 33-34.

[66] 刘凯. 采煤机调高液压系统故障树分析[J]. 煤矿机械, 2020, 41(09): 171-173.

[67] 赵津. 采煤机潜在故障预测和可靠性分析[D]. 太原理工大学，2016.

[68] 刘敏. 基于粗糙集和神经网络的采煤机液压调高系统故障诊断研究[D]. 太原科技大学，2020.

[69] 杨晋玲. 基于信息融合技术的采煤机液压系统故障诊断研究[D]. 中北大学, 2021.

[70] 迟焕磊，袁智，曹琰，等. 基于数字孪生的智能化工作面三维监测技术研究[J]. 煤炭科学技术，2021, 49(10): 153-161.

[71] 孟蕾，田凌. 基于时序数据库的产品数字孪生模型海量动态数据建模方法[J]. 清华大学学报(自然科学版)，2021, 61(11): 1281-1288.

[72] Park J, Kim S, Chio J, et al. Corrigendum to “Frequency energy shift method for bearing fault prognosis using microphone sensor”[Mech. Syst. Signal Process. 147 (2021) 107068][J]. Mechanical Systems and Signal Processing, 2022, (62): 11-20.

[73] 陆宁云，陈闯，姜斌，等. 复杂系统维护策略最新研究进展:从视情维护到预测性维护[J]. 自动化学报，2021, 47(01): 1-17.

[74] 梁兰. 基于数据融合的采煤机液压系统故障诊断研究[D]. 西安：西安科技大学，2016.

[75] 张超. 矿井主通风机状态监测与故障预警系统研发[D]. 西安科技大学, 2020.

[76] 郭宇，杨育. 基于灰色粗糙集与BP神经网络的设备故障预测[J]. 计算机应用研究，2017, 34(09): 2642-2645.

[77] 张雨萌. 数字孪生驱动的矿用设备维修MR辅助指导系统[D]. 西安科技大学，2020.

[78] 郭宇. 基于故障预测的设备预防性维护策略研究及应用[D]. 重庆大学，2017