Neural Network Deployment on Edge via OPC UA Protocol

2022
With the rise of Industry 4.0, the transmission and processing of massive data has become a key technical problem. The OPC UA protocol is considered to be a vital solution for the unification of data transmission protocols. Edge computing services which are physically close to the site are considered to be excellent solutions to reduce transmission delay with low bandwidth occupation. However, the combination of edge computing services and OPC UA is complicated. The extra adjustment of services limits the deployment of advanced algorithm in industry field. To solve the problem, this paper proposes an OPC UA information model for neural network abstraction and instantiation. The Mobile Neural Network is introduced to the framework for rapid deployment of neural networks. Experiments proved that the online deployment, adjustment and execution of neural networks can be realized without modifying the client configuration under the OPC UA framework.
    • Correction
    • Source
    • Cite
    • Save
    18
    References
    0
    Citations
    NaN
    KQI
    []
    Baidu
    map