多模态数据驱动的智能故障诊断方法OA

The multi-modal data can provide more comprehensive and multi-dimensional operation status information of mechanical equipment than the single-modal data in the data-driven intelligent rotating machinery fault diagnosis(RMFD).The method of multi-modal data-driven intelligent fault diagnosis can significantly improve the accuracy and robustness of RMFD.However,the multi-modal data collected by different types of sensors in the operation of rotating machinery equipment are large scale and have significant heterogeneity and complementarities.How to effectively extract and fuse the fault features of different modalities is a key problem to be solved in multi-modal data-driven fault diagnosis.On this basis,a method of multi-modal data-driven intelligent fault diagnosis is proposed.The multimodal data consisting of vibration signals and current signals are constructed into multiple multimodal radius graphs containing multimodal fault features based on the radius neighbor algorithm,so that the model can effectively learn and extract deep-level information of multimodal fault features.The input and output of each layer in the graph sample and aggregate(GraphSAGE)network are weighted and fused to fully capture the potential associa-tions in multi-modal data and improve the expression ability of the model.A series of experiments are carried out to verify the effectiveness of the proposed method,and the results show that the method has high accuracy in fault diagnosis.