Industrial big data model rapid optimization method based on twin network incremental learning

Abstract

The invention discloses an industrial big data model rapid optimization method based on twin network incremental learning, and belongs to the field of deep learning and the field of industrial intelligent manufacturing. The invention provides a model optimization method based on a feed-forward neural network (FNN), a bidirectional long-short-term memory network (Bi-LSTM), a twin network and an attention model (Attention), and aims to solve the problems that a current industrial data model based on deep learning is single in use scene and cannot perform self-optimization and iterative updating by timely utilizing incremental data generated in real time. The invention provides the model optimization method based on the FNN, the bidirectional long-short-term memory network (Bi-LSTM), the twin network and the Attention. Compared with the existing industrial data model, the method has the advantages that the generalization ability of the industrial data model is improved, and the model can be updated and optimized by utilizing incremental data generated in real time in industrial manufacturing in time.

Description

technical field [0001] The invention belongs to the field, and in particular relates to a rapid optimization method for an industrial big data model based on twin network incremental learning. Background technique [0002] With the rapid development of artificial intelligence and Internet of Things technology, industrial sensors, industrial Internet, automatic control systems, ERP, and other information technologies are widely used in the industrial field, and a large number of data related to industrial production activities are collected in real time and stored in the enterprise's In the information system, especially when the production line in the manufacturing enterprise is running at a high speed, a large amount of data is generated on the industrial equipment, and at the same time, the people and computers in the enterprise will generate large-scale data. The analysis and utilization of these data will help to improve production efficiency, improve production process,...

AI Technical Summary

Problems solved by technology

[0005] The applicable scenario is single, and it requires high computing costs to train different models for different models, raw material formulas, and manufacturing processes in the factory before they can be used in actual production;

[0006] With the improvement of production technology and the change of new data, the model should also be optimized to adapt to the new industrial manufacturing. However, the traditional method cannot update the model with the incremental data generated in real time.

Images