Cutting state edge intelligent monitoring method based on online incremental wear evolution model

Abstract

The technology discloses a cutting state edge intelligent monitoring method based on an online incremental wear evolution model. A tool edge intelligent monitoring system is used for acquiring cuttingforce signals, vibration signals and acoustic emission signals in the machining process of machining equipment in real time; a wireless data transmission network is established, a data transmission path is controlled with a domain controller, and finally data is transmitted from a sensing layer to an edge computing layer; the data is preprocessed with a Field Programmable Gate Array (FPGA) hardware system of the edge computing layer to obtain a training sample, an internal topological structure of an online sequential increment extreme learning machine is determined by combining the feature number and the sample number of the training sample, and finally the data acquired by a sensor is transmitted to the trained edge computing layer under the same working condition; a real-time monitoring result is sent to a cloud computing layer by utilizing an SDN technology layer in the tool edge intelligent monitoring system; and a tool wear monitoring result is applied by using an application service in the cloud computing layer. The technology has high real-time performance, good accuracy and strong data processing capability.

Description

technical field [0001] The technology relates to the field of tool wear monitoring of processing equipment, in particular to an intelligent monitoring method for cutting state edges based on an online incremental wear evolution model. Background technique [0002] In the field of mechanical processing, the amount of tool wear is of great significance, which directly affects the remaining life of the tool and the size and surface roughness of the workpiece. However, in the cutting process, the tool is inevitably worn. So far, the methods of monitoring tool wear are mainly divided into direct method and indirect method. The direct method refers to the contact method and the optical image method. However, due to the limitation of the test conditions, the direct method needs to stop the processing equipment at different time stages, and The required infrastructure is complex and difficult to operate; the indirect method refers to monitoring the cutting force signal, tool vibrati...

AI Technical Summary

Problems solved by technology

However, because the current and voltage are easily affected by external factors, the monitoring results cannot accurately represent the real-time wear status of the tool, and the changes in current and voltage can only determine whether the tool has been worn, but not the real-time wear of the tool.

[0008] At present, the existing technologies of the indirect method all use batch learning methods and offline training methods. In order to meet the requirements of online monitoring, the previous learning results need to be discarded, and it is necessary to re-learn and train based on the newly added data and all past data. Learning all the data will consume a lot of time and space resources; as the scale of data continues to increase, the demand for time and space will also increase rapidly, and eventually the speed of learning will not even catch up with the speed of data update; in practical applications, It is usually impossible to obtain all the training samples at once, and various sensory data representing the wear state of the tool are often obtained gradually over time, and the information reflected by the samples may also change over time; these problems seriously affect the machining process On-Line Requirements of Medium Tool Wear Monitoring System

[0009] In addition, in the existing technology of the indirect method, the data processing in the processing process is completed in the server or data center. Due to the large amount of data generated in the processing process, the transmitted data is slow and affects the calculation of the data center. The speed increases the time of data processing, and fails to meet the real-time requirements of the tool wear monitoring system in the machining process.

In response to this problem, this technology plans to design an intelligent monitoring method for the edge of the cutting state based on the online incremental wear evolution model. This system uses the online sequential incremental extreme learning machine model and the edge computing method to effectively solve the problem of tool wear monitoring in the current machining process. The problem of low real-time performance and insufficient accuracy of the system

Images