Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5469861 | Procedia CIRP | 2016 | 6 Pages |
Abstract
A model to track the cutting usage has been developed. The application domain of the model is milling in the woodwork industry. The model involves the capturing of raw data from three different sources; the machine tool control, the Enterprise-Resource-Planning system and the tool management system. The data is analysed, the cutting distance is calculated and stored along with the data of the tool usage parameter settings. In order to predict the remaining tool life it is essential to know the actual tool usage. The usage is based on the tool life distance which is calculated by the movement data of the axis and various parameters of the numerical control system. Input for the calculation is the cutting path, the tool usage data and the end of life due to tool wear. With this information it is possible to forecast the estimated tool life by using machine learning algorithms. This forecast algorithm will predict more accurate results after each learning cycle. The developed model has been implemented as a smart service. The tool wear predictions are used for improvement of machine availability. Tool changes can be done in advance of large orders and machining can be operated with less supervision. Furthermore the shipping of tools between the manufacturers, the customer and the refurbishment can be optimised. Premature wear due to wrong settings or human error can be avoided. The system ensures a reliable exchange of information without paper work via the internet across company boarders for the whole life cycle of each tool.
Related Topics
Physical Sciences and Engineering
Engineering
Industrial and Manufacturing Engineering
Authors
Juergen Lenz, Dominik Brenner, Engelbert Westkaemper,