Conceptual Framework of Smart Factory Based on OPC UA and LSTM Encoder-Decoder

Abstract

OPC UA (Open Platform Communications Unified Architecture) is key for multi-heterogeneous software and hardware information integration. Some studies have proposed to apply OPC UA to smartfactory information integration, but most tend to propose concepts and lack application research. While, LSTM Encoder-Decoder (Long Short-Term Memory based Encoder-decoder) is an unsupervised learning method not requiring labeled data. It has good performance in time series data analysis, but few researches relate it with smart factory predictive analysis. This paper proposed a conceptual framework of smart factory based on OPC UA and LSTM Encoder-Decoder technology. Specifically, first, the theories are discussed including the general architecture of smart factory, information integration architecture based on OPC UA and LSTM Encoder-decoder model for smart factory predictive analysis. Then, we introduced the simulation result analysis. Finally, the conclusion was given.