Research On Edge Computing And Application For Intelligent Equipment

The traditional intelligent equipment for automatic control is based on signals,and there is a problem of extensive control strategies.The intelligent equipment based on cloud computing is a self-regulating system based on data,but there is a problem that the transmission distance between the cloud computing center and the edge equipment at the factory end is too far,resulting in insufficient control effectiveness.With the advent of the 5G era,the amount of data circulating in the network will increase.For factories,the acceleration of transmission speed will lead to a surge in the amount of data in the same time,so the requirements for network bandwidth will become higher and higher,and cloud storage will also be added to store massive amounts of data.In recent years,edge computing technology has become increasingly mature.Edge computing is a terminal device between the cloud and the edge device at the physical level.Because of its own computing power,it is closer to the edge device than the cloud server.It can provide computing power to edge devices faster and more conveniently,so that industrial equipment can make quick response and processing without connecting to the cloud,and provides new processing control solutions for intelligent equipment.This paper analyzes the edge computing technology by comparing cloud computing,and designs an intelligent equipment architecture based on edge computing.By analyzing the needs of the intelligent equipment,the three modules of data acquisition module,data communication module and data processing module are designed.The architecture is divided into three parts: device layer,edge layer and cloud platform layer.In the transformation of the traditional pumps of the sewage treatment plant,the previously designed intelligent equipment based on edge computing was specifically implemented.By reserving sensor ports to enhance the scalability of the architecture acquisition module,to ensure the future expansion needs of the factory.Because the connection methods between the edge layer and the device layer and between the edge layer are different,different data communication schemes are used separately.The data collected by the device layer is transmitted to the edge layer through fieldbus technology.A coprocessor is added to communicate with each node;the internal devices at the edge layer complete network communication through Ethernet.The data uploaded by the device layer adopts time series-based data anomaly detection to check the data integrity and accuracy,and uses FFT-based data processing to preprocess the original data,and retains valid data through data cleaning.The principal component analysis algorithm is used to reduce the dimensionality of the data,extract the characteristic values of the fault type,and use the SPE statistics and T2 statistics to determine whether the device is faulty.Finally,input the data into the machine learning algorithm library to train the energy optimization algorithm model.After the artificial intelligence energy optimization algorithm is developed,the complete control program is updated to the intelligent energy-saving terminal to realize energy-saving control of the pump station.