Research on Manufacturing Data Space Storage based on Data Space

With the globalization of the manufacturing supply chain, the data of the manufacturing shows the big data characteristics of diversity, multi -source, and large data. The amount of data around manufacturing, industry, and people’s lives has gradually increased, making manufacturing and people enter the data space. This paper divides the data in the data space into three categories: structured data, non - structured data, and semi -structured data, and proposes an extract framework of the data space. And then this paper proposes the processing process of data pre-processing in the data space. In order to confirm the feasibility of the model in this paper, we select the load prediction in the power industry for verification. We extract relevant non -structured weather data and characterize the data, combine with structured data for extraction of load sequences, and build a LSTM neural network model to forecasting the electricity usage. It is proved that the prediction accuracy is high, which confirms the effectiveness of the concept and model proposed in this paper.