Abstract: A data acquisition architecture that integrates unit devices into a unified OPC UA architecture for communication is designed in order to address the problem of data acquisition difficulties and response time extension, etc. during a large number of acquired communication protocols data for smart made units of dropper pre-assembly of high-speed railway is combined, then to be centralized and transferred to the cloud for analysis. An end-to-end motor fault real-time diagnosis method applied to the edge field has been proposed by combining the characteristics of motor vibration data in the cloud after the collected data is preprocessed at the edge nearby and then coordinated to the cloud. The application of the proposed method in the cantilever pre-assembly production line has been verified, and the results show that the data acquisition and fault analysis method is universal and efficient.