Characterising Acoustic Emission Signals for the Online Monitoring of a Fluid Magnetic Abrasives Finishing Process

To implement an automated fluid magnetic abrasive (FMA) finishing process, an online monitoring scheme is proposed based on characterising acoustic emission (AE) signals in this paper. According to the material removal mechanisms during the FMA finishing process, the AE generation and characteristics are predicted analytically to be dominated by the interactions between the surface asperities and the abrasive particles. Moreover, the interactions and corresponding AE events will become weaker as the finishing process progresses and the surface becomes smoother. Experimental studies show that the amplitude and the occurrence rate of continuous AE waves and intermediate bursts reduce gradually with the progression of the finishing process. Based on these features, root mean squared (RMS) values and burst occurrence rates (BOR), being of the lowest computational requirements, are suggested as online monitoring parameters for an automated and intelligent finishing in FMA manufacturing. The proposed method is verified experimentally, showing that the RMS values are highly consistent with measured surface roughness values, which confirms the dynamic mechanisms between the FMA finishing and AE generation sources examined.