In-process detection of grinding burn using machine learning

The improvement of industrial grinding processes is driven by the objective to reduce process time and costs while maintaining required workpiece quality characteristics. One of several limiting factors is grinding burn. Usually applied techniques for workpiece burn are conducted often only for selected parts and can be time consuming. This study presents a new approach for grinding burn detection realized for each ground part under near-production conditions. Based on the in-process measurement of acoustic emission, spindle electric current, and power signals, time-frequency transforms are conducted to derive almost 900 statistical features as an input for machine learning algorithms. Using genetic programming, an optimized combination between feature selector and classifier is determined to detect grinding burn. The application of the approach results in a high classification accuracy of about 99% for the binary problem and more than 98% for the multi-classdetection case, respectively.

Grinding Burns Testing — New NDT Method? Do You Need Qualified Specialists for This?

The highly loaded components used in machine building, automotive and aviation must have no thermal damage during grinding, the so-called grinding burn. In less loaded components, the grinding burn shall not exceed a certain level agreed between the manufacturer and the customer. In industrial practice, etching, Barkhausen-Rausch methods and eddy current are used to detect burns. To reliably perform inspection according to any of the methods, it is necessary that the flaw detection experts have sufficient knowledge and skills. This is particularly relevant for the multi-stage chemical etching process. In 2015, at the request of Turkish partners, IMQ developed the course "Inspection of grinding burns using chemical etching" in accordance with ISO 14104 and AMS 2649. Since then, more than 70 participants from several European countries have successfully completed this two-day course. In cooperation with DGZfP Ausbildung und Training GmbH, IMQ has developed a new course for the training of flaw detectors in the burns inspection during grinding. In addition to the etching method, the course included the eddy current method and the Barkhausen method, as well as liquid penetrant testing and magnetic powder testing. The content of the course, practical exercises and elements of the final exam in theory and practice are presented.

Scalogram-Based Instantaneous Features of Acoustic Emission in Grinding Burn Detection

Time-frequency methods are effective tools in identifying the frequency content of a signal and revealing its time-variant features. This paper presents the use of instantaneous features (i.e. instantaneous energy and signal phase) of acoustic emission (AE) in the detection of thermal damage to the workpiece in grinding. Both the instantaneous energy and mean frequency are obtained using the low-order frequency moments of a scalogram. While the zero-order frequency moment yields the instantaneous energy, the first-order frequency moment gives the instantaneous frequency by which the signal phase is recovered. The grinding process is monitored using acoustic emission for various operating conditions, including the regular grinding, grinding at a higher cutting speed and larger infeed, and small dressing depth of cut. It has been found that both the instantaneous energy and phase deviation indicate the presence of burn damage and serve as robust and reliable indicators, providing a basis for detecting the grinding burn.

Scalogram-Based Instantaneous Features of Acoustic Emission in Grinding Burn Detection

Time-frequency methods are effective tools in identifying the frequency content of a signal and revealing its time-variant features. This paper presents the use of instantaneous features (i.e. instantaneous energy and signal phase) of acoustic emission (AE) in the detection of thermal damage to the workpiece in grinding. Both the instantaneous energy and mean frequency are obtained using the low-order frequency moments of a scalogram. While the zero-order frequency moment yields the instantaneous energy, the first-order frequency moment gives the instantaneous frequency by which the signal phase is recovered. The grinding process is monitored using acoustic emission for various operating conditions, including the regular grinding, grinding at a higher cutting speed and larger infeed, and small dressing depth of cut. It has been found that both the instantaneous energy and phase deviation indicate the presence of burn damage and serve as robust and reliable indicators, providing a basis for detecting the grinding burn.