Design And Implementation Of Industrial Data Acquisition And Processing System In Cloud-edge Collaborative Environment

At present,a new round of technological revolution and industrial transformation are accelerating the transformation of economic development mode.The manufacturing industry is facing major adjustments.The rapid development of new generation information technologies such as big data,cloud computing,and edge computing provides opportunities for the transformation and upgrading of traditional manufacturing enterprises.With the deep integration of a new generation of information technology and manufacturing,the amount of industrial data is increasing exponentially.In order to find the laws contained in the massive amount of industrial data and help the transformation and upgrading of manufacturing enterprises,data collection and data mining system are developed based on the Industrial Internet of Things platform.This system has important theoretical significance and practical value.This paper takes intelligent manufacturing companies as the research object,and aims to dig out the laws contained in equipment operating data.Using cloud-edge collaboration technology and Industrial Internet of Things platform to complete edge device data collection,edge data preprocessing,cloud platform equipment fault diagnosis modeling analysis,industrial data management and other functional design and implementation,which can help manufacturing enterprises to transform and improve their production efficiency.Firstly,based on existing problems such as a wide variety of collected data and high transmission delays on cloud platforms,an Industrial Internet of Things implementation architecture in a cloud-edge collaborative environment is designed based on edge computing technology,covering all functional requirements from edge data collection to cloud platform data mining.Secondly,connecting production equipment in the manufacturing company through the DNC network and the collected data is modeled by category to obtain five data models including equipment information data model,staff information data model,and production information data model,realize short-distance transmission through industrial fieldbus and 5G wireless network,and store the data directly into the edge storage server.Thirdly,the collected industrial data is preprocessed at the edge gateway,and the Isolation Forest algorithm and the Simulated Annealing algorithm are combined to detect anomalies.Aiming at the problem of poor stability caused by the artificial introduction of random variables in the Isolation Forest algorithm,the SA algorithm is used to select an isolation tree with better adaptability to generate an isolation forest.Many simulation results show that the SA-i Forest algorithm reduces redundancy and improves reliability,and the accuracy of novelty detection is also significantly improved.Established a fault diagnosis evaluation model on the Industrial Internet of Things cloud platform based on the XGBoost algorithm,and compared with the traditional Bayes algorithm and GBDT algorithm.The quantified performance of the chaotic matrix evaluation model proves that the XGBoost algorithm has superiority in equipment novelty detection.Finally,through the software design and feasibility analysis of the industrial data management system,an industrial data management platform based on Spring Boot was developed and deployed to the cloud server.The industrial data management system realizes the functions of industrial data visualization,production equipment management,warehousing logistics management,and so on.It also applies the fault diagnosis evaluation model to the system,and realizes the intelligent management and control of alarm events.Through the system function test,it is proved that the various modules of the system can operate stably and can meet the needs of intelligent manufacturing enterprises.