An Industry 4.0 example: real-time quality control for steel-based mass production using Machine Learning on non-invasive sensor data

2022

Insufficient steel quality in mass production can cause extremely costly damage to tooling, production downtimes and low quality products. Automatic, fast and cheap strategies to estimate essential material properties for quality control, risk mitigation and the prediction of faults are highly desirable. In this work we analyse a high throughput production line of steel-based products. Currently, the material quality is checked using manual destructive testing, which is slow, wasteful and covers only a tiny fraction of the material. To achieve complete testing coverage our industrial collaborator developed a contactless, non-invasive, electromagnetic sensor to measure all material during production in real-time. Our contribution is three-fold: 1) We show in a controlled experiment that the sensor can distinguish steel with deliberately altered properties. 2) During several months of production 48 steel coils were fully measured non-invasively and additional destructive tests were conducted on samples taken from them to serve as ground truth. A linear model is fitted to predict from the non-invasive measurements two key material properties (yield strength and tensile strength) that normally have to be obtained by destructive tests. The performance is evaluated in leave-one-coil-out cross-validation. 3) The resulting model is used to analyse the material properties and the relationship with reported product faults on real production data of approximately 108 km of processed material measured with the non-invasive sensor. The model achieves an excellent performance (F3-score of 0.95) predicting material running out of specifications for the tensile strength. In a second controlled experiment one coil suspected of material faults was sampled 18 times over its full length and repeated non-invasive as well as destructive testing was performed to analyse the relationship between both measurement types in a situation where also product faults and problems during production are expected to occur. On this coil the model predictions demonstrate that material properties are indeed out of specification near the point for which the products made from the neighbouring coil exhibited faults during production. The combination of model predictions and logged product faults shows that if a significant percentage of estimated yield stress values is out of specification, the risk of product faults is high. Our analysis demonstrates promising directions for real-time quality control, risk monitoring and fault detection.

Link

@inproceedings{Straat2022,
title = {An Industry 4.0 example: real-time quality control for steel-based mass production using Machine Learning on non-invasive sensor data},
author = {Michiel Straat and  Kevin Koster and Nick Goet and Kerstin Bunte},
month = {7},
organization = {IEEE},
editor = {"},
booktitle = {2022 International Joint Conference on Neural Networks (IJCNN)},
year = {2022},
publisher = {},
address = {Padua, Italy},
pages = {01--08},
abstract = {Insufficient steel quality in mass production can cause extremely costly damage to tooling, production downtimes and low quality products. Automatic, fast and cheap strategies to estimate essential material properties for quality control, risk mitigation and the prediction of faults are highly desirable. In this work we analyse a high throughput production line of steel-based products. Currently, the material quality is checked using manual destructive testing, which is slow, wasteful and covers only a tiny fraction of the material. To achieve complete testing coverage our industrial collaborator developed a contactless, non-invasive, electromagnetic sensor to measure all material during production in real-time. Our contribution is three-fold: 1) We show in a controlled experiment that the sensor can distinguish steel with deliberately altered properties. 2) During several months of production 48 steel coils were fully measured non-invasively and additional destructive tests were conducted on samples taken from them to serve as ground truth. A linear model is fitted to predict from the non-invasive measurements two key material properties (yield strength and tensile strength) that normally have to be obtained by destructive tests. The performance is evaluated in leave-one-coil-out cross-validation. 3) The resulting model is used to analyse the material properties and the relationship with reported product faults on real production data of approximately 108 km of processed material measured with the non-invasive sensor. The model achieves an excellent performance (F3-score of 0.95) predicting material running out of specifications for the tensile strength. In a second controlled experiment one coil suspected of material faults was sampled 18 times over its full length and repeated non-invasive as well as destructive testing was performed to analyse the relationship between both measurement types in a situation where also product faults and problems during production are expected to occur. On this coil the model predictions demonstrate that material properties are indeed out of specification near the point for which the products made from the neighbouring coil exhibited faults during production. The combination of model predictions and logged product faults shows that if a significant percentage of estimated yield stress values is out of specification, the risk of product faults is high. Our analysis demonstrates promising directions for real-time quality control, risk monitoring and fault detection.},
doi = {10.1109/IJCNN55064.2022.9892432},
url = {},
isbn = {"},
}