An edge computing gateway for device pre-maintenance

Abstract

The invention provides an edge computing gateway for equipment pre-maintenance. The edge computing gateway comprises a protocol acquisition module, an edge computing module, an output control module, an HMI man-machine interaction module and an MQTT communication module, the protocol acquisition module is used for acquiring data information of the Internet of Things equipment; the edge calculation module is used for training a neural network algorithm model according to the data information acquired by the protocol acquisition module, inputting the acquired data information into the neural network algorithm model for operation after the training is completed, judging whether the equipment works in a normal interval or not according to the acquired data information, and generating prediction alarm information when the equipment is judged to be abnormal; the HMI man-machine interaction module is used for setting operation parameters of the Internet of Things equipment through the touch display screen; and the MQTT communication module is used for communicating with a server. The invention can be installed on an industrial site nearby, the transmission data volume is reduced, and the data receiving concurrency of the server is reduced for timely early warning.

Description

technical field [0001] The invention relates to an edge computing gateway, in particular to an edge computing gateway for equipment pre-maintenance. Background technique [0002] Most of the existing industrial gateways are industrial protocol gateways. For example, the equipment data on the industrial site is collected into the gateway through a certain protocol, and then uploaded to the server through a certain network protocol (such as MODBUSTCP). The data packet is transparently transmitted to the server without any processing. Some gateways will do some data analysis, but the functions are relatively simple. Generally, they do not support direct docking with mainstream IoT platforms, and cannot meet the current data collection needs of the industrial Internet industry. Moreover, they do not support edge computing functions and cannot analyze data at the gateway. Real-time processing, and effective intervention on equipment in real time according to the processing resul...

AI Technical Summary

Problems solved by technology

[0002] Most of the existing industrial gateways are industrial protocol gateways. For example, the equipment data on the industrial site is collected into the gateway through a certain protocol, and then uploaded to the server through a certain network protocol (such as MODBUSTCP). The data packet is transparently transmitted to the server without any processing

Some gateways will do some data analysis, but the functions are relatively simple. Generally, they do not support direct docking with mainstream IoT platforms, and cannot meet the current data collection needs of the industrial Internet industry. Moreover, they do not support edge computing functions and cannot analyze data at the gateway. Real-time processing, and effective intervention on equipment in real time according to the processing results

[0003] Industrial Internet predictive maintenance applications need to collect massive sensory data in real time. Most of these source data are not important, and it is impractical to transmit such a large amount of data to the cloud

Images