Laser Cut Interruption Detection from Small Images by Using Convolutional Neural Network.

Benedikt Adelmann,Ralf Hellmann,Max Schleier

doi:10.3390/s21020655

Benedikt Adelmann, Ralf Hellmann + Show 1 more

Open Access

https://doi.org/10.3390/s21020655

Copy DOI

Abstract

In this publication, we use a small convolutional neural network to detect cut interruptions during laser cutting from single images of a high-speed camera. A camera takes images without additional illumination at a resolution of 32 × 64 pixels from cutting steel sheets of varying thicknesses with different laser parameter combinations and classifies them into cuts and cut interruptions. After a short learning period of five epochs on a certain sheet thickness, the images are classified with a low error rate of 0.05%. The use of color images reveals slight advantages with lower error rates over greyscale images, since, during cut interruptions, the image color changes towards blue. A training set on all sheet thicknesses in one network results in tests error rates below 0.1%. This low error rate and the short calculation time of 120 µs on a standard CPU makes the system industrially applicable.

Highlights

Cutting metals by fiber or disk lasers is nowadays a standard production process in the modern industry
26% showed cut interruptions, three-mm sheets at 3000-W laser power with 105,000 images overall was used for the trainwhile the others showed complete cuts
We used a small convolutional neural network with a calculation time of 120 μs to classify images from the processing zone during laser cutting into complete cuts or cut interruptions

Summary

Introduction

Cutting metals by fiber or disk lasers is nowadays a standard production process in the modern industry. While available laser powers rise continuously up to 30 kW or higher [1], cutting 100-mm-thick sheets is possible [2]. Due to the general trend of higher automation, with the result of unmanned machines and the seamless combination of laser cutting machines with bending, separation or welding technologies, high and reliable quality of cuts are necessary to avoid downtime or damaging subsequent machine steps in such combined process chains. To obtain high-quality cuts, the process parameters, such as laser power, feed rate, gas pressure, working distance of the nozzle and focus position, respectively, must be selected appropriately. Imprecise process parameters and typical disturbance values like thermal lenses, unclean optics, damaged gas nozzles, gas pressure fluctuations and the variations of material properties may lead to poor-quality and, nonconforming products. To ensure a high quality, an online quality monitoring system would be the best choice, which allows a quick response, reduces downtime or cost-extensive rework and saves material

Methods

Results

Discussion

Conclusion