Data analysis system for industrial site side cloud collaboration and implementation method

Abstract

The invention provides a data analysis system for industrial site side cloud collaboration and an implementation method thereof. An edge-cloud collaborative architecture is adopted in a data analysis mode, and the whole system is composed of a data acquisition node, an edge data processing node and a cloud service center, wherein the data acquisition node acquires industrial field data in real time; the edge data processing node carries a distributed artificial intelligence model, pre-processes and analyzes the industrial data in real time to obtain an analysis result, displays the result in real time, and sends the pre-processed data and the analysis result to the cloud; and the cloud service center carries a big data artificial intelligence model, summarizes analysis results and data of the distributed edge nodes, optimizes and updates the distributed artificial intelligence model of the edge end through deep learning and incremental learning, and displays the analysis results to the user. Through cooperative work of the edge end and the cloud end, high-efficiency, high-value and intelligent industrial field data analysis can be realized.

Description

technical field [0001] The invention belongs to the technical field of industrial Internet, and specifically relates to a data analysis system and implementation method for edge-cloud collaboration on industrial sites, including the overall architecture of the system, construction methods of each end, and collaboration methods, and is applicable to the background of "smart manufacturing" Intelligent data analysis of industrial sites. Background technique [0002] With the rapid development of wireless communication technology and the Internet, the Internet of Things came into being. It connects various physical quantity sensor devices into a network through prescribed information transmission rules, helping us realize the state monitoring and environmental awareness of management objects, and improve our Ability to control extended objects. At present, the scale of industrial production continues to expand, the level of intelligence in the manufacturing industry continues t...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the above-mentioned problems existing in the prior art, the present invention discloses a data analysis system for edge-cloud collaboration in industrial sites and its implementation method. The edge-end data processing nodes are equipped with distributed artificial intelligence models to complete the analysis of industrial site data. Preprocessing and real-time analysis, and uploading the processed data and analysis results to the cloud solves the problems of data delay and result sharing; the cloud relies on its powerful computing power to summarize the data uploaded by multiple nodes, through incremental learning and The deep learning method optimizes and updates the distributed model of the edge in a timely manner, breaking the information islands between each edge, and also solving the problems of limited computing resources at the edge and inability to update the model in time

Images