Real-Time Identification of Incipient Surface Morphology Variations in Ultraprecision Machining Process

Abstract

Real-time monitoring and control of surface morphology variations in their incipient stages are vital for assuring nanometric range finish in the ultraprecision machining (UPM) process. A real-time monitoring approach, based on predicting and updating the process states from sensor signals (using advanced neural networks (NNs) and Bayesian analysis) is reported for detecting the incipient surface variations in UPM. An ultraprecision diamond turning machine is instrumented with three miniature accelerometers, a three-axis piezoelectric dynamometer, andan acoustic emission (AE) sensor for process monitoring. The machine tool is used for face-turning aluminum 6061 discs to a surface finish (RJ in the range of 15-25 nm. While the sensor signals (especially the vibration signal in the feed direction) are sensitive to surface variations, the extraneous noise from the environment, machine elements, and sensing system prevents direct use of raw signal patterns for early detection of surface variations. Also, nonlinear and time-varying nature of the process dynamics does not lend conventional statistical process monitoring techniques suitable for characterizing UPM-machined surfaces. Consequently, instead of just monitoring the raw sensor signal patterns, the nonlinear process dynamics wherefrom the signal evolves are more effectively captured using a recurrent predictor neural network (RPNN). The parameters of the RPNN (weights and biases) serve as the surrogates of the process states, which are updated in real-time, based on measured sensor signals using a Bayesian particle filter (PF) technique. We show that the PF-updated RPNN can effectively capture the complex signal evolution patterns. We use a mean-shift statistic, estimated from the PF-estimated surrogate states, to detect surface variationinduced changes in the process dynamics. Experimental investigations show that variations in surface characteristics can be detected within 15 ms of their inception using the present approach, as opposed to 30 ms or higher with the conventional statistical change detection methods tested.

Dynamic network monitoring and control of in situ image profiles from ultraprecision machining and biomanufacturing processes

“…Furthermore, we propose an NGLR chart to monitor the multivariate vector of network community statistics. For a stream of m image profiles, we have the data sample of (y (1) , y (2) , … , y (i) , … , y (m) ). Under the null hypothesis H 0 , all m feature vectors are from in-control distribution and the likelihood function is…”

“…As shown in Figure A, ultraprecision machining (UPM) is equipped with air‐bearing spindles and diamond tools to produce optical surface finishes. The 2D and 3D images are often collected to inspect the quality of surface finishes, which is closely pertinent to machine setup and the operation of cutting tools . Figure B shows time‐varying 3D (4D) images of cardiac electrical waves captured through optimal mapping techniques.…”

“…The 2D and 3D images are often collected to inspect the quality of surface finishes, which is closely pertinent to machine setup and the operation of cutting tools. 1 Figure 1B shows time-varying 3D (4D) images of cardiac electrical waves captured through optimal mapping techniques. Such dynamic images are critical to the study of the pathology of heart diseases.…”

Pairwise Critical Point Detection Using Torque Signals in Threaded Pipe Connection Processes

“…Shao et al developed a quadratic classifier and extracted space and frequency domain features of cross-sectional profiles on tool surfaces for tool wear monitoring in the battery manufacturing process [23]. Rao et al used a recurrent predictor neural network to compactly capture heterogeneous vibration signals to detect changes induced by surface variations in ultraprecision machining process monitoring [24]. Wu et al developed an acoustic monitoring method based on two kinds of cavitation noises to monitor the micro/nanoparticle dispersion status in aqueous liquid [25].…”

Possibilities of the Generation of Hardened Steel Parts With Defined Topographic Characteristics of the Machined Surfaces

“…Later, Waikar and Guo [10] compared 3D surface topographies produced by hard turning and grinding of AISI 52100 steel of 61-62 HRC (using CBN tools and AI2O3 wheels). Rao et al [11] have investigated how surface morphology varies in ultraprecision machining. The 2D and 3D comparison, more oriented on bearing area parameters, related to PHT and belt grinding is provided by Grzesik et al [9,12], Nowadays, the technological shifts in surface metrology allow the surface features generated by modern manu facturing processes (including hard part machining) to be charac terized with higher accuracy using a number of the field parameters (.S'-parameters and V-parameters sets) [13].…”