基于迁移学习的采煤机摇臂传动故障研究

针对传统故障诊断方法无法有效识别并自动分类实际工况中采煤机摇臂传动故障多的非线性、非平稳信号，提出一种基于迁移学习的采煤机摇臂传动故障诊断模型。基于迁移学习思想，构建基于深度迁移学习的采煤机摇臂传动故障诊断模型；采用多标签分类及sigmoid函数，对模型进行改进，实现对采煤机摇臂传动复合故障的识别与分类；最后，通过仿真实验验证了改进模型性能，并对比了提出模型与传统智能故障诊断模型DCNN、SVM、LSTM、CNN在迁移任务中的分类准确率。结果表明，相较于传统智能故障诊断模型，基于深度迁移学习的采煤机摇臂传动故障诊断模型具有更高的诊断精度，且收敛速度更快，可提高采煤机摇臂传动系统的工作可靠性。

Study on transmission failure of shearer rocker arm based on transfer learning

Aiming at the traditional fault diagnosis methods can not effectively identify and automatically classify the non-linear and non-stationary signals of shearer rocker drive faults in actual working conditions，a fault diagnosis model of shearer rocker drive based on transfer learning was proposed.Based on the idea of transfer learning，the fault diagnosis model of shearer rocker arm transmission based on deep transfer learning was constructed.Multi label classification and sigmoid function were used to improve the model and realize the recognition and classification of shearer rocker transmission composite fault.Finally，the performance of the improved model was verified by simulation experiments，and the classification accuracy of the proposed model was compared with the traditional intelligent fault diagnosis models DCNN，SVM，LSTM and CNN in the migration task.The results showed that compared with the traditional intelligent fault diagnosis model，the fault diagnosis model based on deep migration learning has higher diagnostic accuracy and faster convergence speed，which can improve the reliability of the coal cutter rocker drive system