Real-time simulation of neural network classifications from characteristics emitted by acoustic emission during horizontal single grit scratch tests

With an increase of exotic materials and the machine process often associated with obtaining tight tolerances is that of grinding. There is an increased need in understanding the fundamental mechanics to be able to accurately model such material and grit interactions. For this reason the unit event of material interaction in grinding are investigated. Where three phenomenon’s are involved namely: rubbing, ploughing and cutting. Ploughing and rubbing essentially mean the energy is being applied less efficiently in terms of material removal. Such phenomenon usually occurs before or after cutting. Based on this distinction it is important to identify the effects of these different phenomena experienced during grinding. Two acoustic emission (AE) sensors were used to extract the very fast, transient material and grit interaction to correlate with the measured material surface. Accurate material surface profile measurements of the cut groove were made using the Fogale Photomap Profiler which enables the comparison between the corresponding AE signal scratch data. Short-time Fourier transforms and filtration techniques ensured the translation and salient components for identification were ready for classification ensuring the distinct levels between the three different Grit (SG) phenomenons. Neural networks (NNs) was used to classify and verify the demarcation of SG phenomena. After the cutting, ploughing and rubbing gave a high confidence in terms of classification accuracy. To map the results from the unit/micro to the multi/macro event both 1 $$\upmu $$ μ m and 0.1 mm grinding test data were applied to the NN for classification. Interesting output results correlated for both classifiers signifying a distinction that there is more cutting utilisation than both ploughing and rubbing as the interaction between grit and workpiece become more involved (measured depth of cut increases). Such findings were then realised into a Simulink model as a potential control system for industrial purposes and a potential model for mapping the micro and macro mechanics seen in grinding technologies.