Abstract: In view of the difficult problems of condition monitoring and maintenance of shearers in poor working environment, combined with the high-fidelity behavior simulation characteristics of digital twin and the powerful data mining ability of deep learning,a prediction method of shearer health status driven by the fusion of digital twin and deep learning was proposed. The digital twin of shearers was constructed based on multi-physical parameters of physical space, and the health state early prognosis of shearers was realized by visual display and analysis in virtual space. A prediction model of RUL of shearer key parts was established based on deep learning, and the online RUL prediction of parts driven by real-time monitoring data was realized. The prediction of shearer health was obtained based on the status of digital twin and the value of RUL. Finally, the effectiveness of the method was verified by experiments, which provides a new idea for the monitoring and management of the health status of the shearers.

Key words: digital twin, deep learning, shearer, health prediction；remaining useful life(RUL) prediction

摘要： 针对处于恶劣工作环境的采煤机状态预测与维护困难的问题，结合数字孪生高逼真度行为仿真特性和深度学习强大的数据挖掘能力，提出数字孪生与深度学习融合驱动的采煤机健康状态预测方法。基于物理空间多物理参数构建采煤机数字孪生体，通过在虚拟空间的可视化展示与分析实现健康状态预判；建立基于深度学习的采煤机关键零件剩余寿命预测模型，实现实时监测数据驱动下的零件剩余寿命的在线预测；综合数字孪生体状态和剩余寿命值,实现采煤机健康状态预测。通过试验验证了该方法的有效性，为采煤机健康状态预测与管理提供新思路。

关键词: 数字孪生, 深度学习, 采煤机, 健康预测, 剩余寿命预测