Automatic grinding burn recognition based on time-frequency analysis and convolutional neural networks
- PDF / 3,349,346 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 93 Downloads / 173 Views
ORIGINAL ARTICLE
Automatic grinding burn recognition based on time-frequency analysis and convolutional neural networks Henrique Butzlaff Hübner 1
&
Marcus Antônio Viana Duarte 1 & Rosemar Batista da Silva 1
Received: 15 April 2020 / Accepted: 9 August 2020 # Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract The grinding process is employed to provide a high-quality surface finish and tight dimensional tolerances to the manufactured components. However, it has the disadvantage of generating a large amount of heat during machining that is mostly transferred to the workpiece when employing conventional grinding wheels, which makes it highly susceptible to thermal damage. In terms of the different thermal damage associated with this process, grinding burn deserves special attention, as it affects the aesthetic aspect of the machined components. Since there is an increasing demand for productivity and high-quality products, the use of systems to monitor grinding burn becomes crucial when global competitiveness is in evidence. In this study, a novel approach, based on time-frequency images of acoustic emission signals and convolutional neural networks was proposed to monitor grinding burn. Experimental data were obtained from grinding tests on N2711 grade steel under different cutting conditions. Three different time-frequency analyses, including the short-time Fourier transform, the continuous wavelet transform, and the Hilbert-Huang transform, were used to generate the images that served as input for the CNN models. Through the proposed approach, grinding burn was successfully recognized, as the highest accuracy obtained by the models was 99.4% on the test dataset. This result is superior when considering those reported in the literature, in which conventional machine learning techniques are employed for grinding burn monitoring. Keywords Acoustic emission . Convolutional neural network . Grinding burn . Process monitoring . Time-frequency analysis
1 Introduction Grinding is one of the most widely used abrasion machining processes in the metal-working industry and is usually performed as one of the latter machining operations to be employed during the manufacture of a component [1]. According to Teixeira et al. [2], the main advantages of this process are the possibility of obtaining high-quality surface finishing and high dimensional accuracy. In comparison with other machining processes, such as turning and milling, for example, a major disadvantage associated with the grinding process is the consumption of large
* Henrique Butzlaff Hübner [email protected] 1
School of Mechanical Engineering, Federal University of Uberlândia, Uberlândia, Minas Gerais 38408-100, Brazil
amounts of energy per unit of material removed, i.e., specific energy. This energy is almost entirely dissipated in the form of heat and mostly directed to the workpiece due to the small chip sections generated as a result of the very low values of depth of cut, as well as the refractory properties of conventional grinding
Data Loading...
