Chapter 11 Variants of high speed machining
Download 0.88 Mb. View original pdf
|
10980-Camera ready paper
| 286 High speed machining The intelligent manufacturing method, cutting parameters in recent years, tool path determination, tool wear track, and surface quality by utilizing artificial neural network (ANN) methods such as optimization are used in studies. In the study by Hassan et al., a general tool condition monitoring approach for the development of intelligent real-time tool condition monitoring systems is presented [8] , which is (1) sensitive to changes in team conditions, (2) insensitive to cutting conditions and adaptive control environments, (c) has a high level of decision accuracy using minimum learning efforts, and (4) performing signal processing and decision making within a suitable time frame. In order to apply these systems to complex processing processes, studies must be conducted to create a self-learning module that can address the high dynamics and variations of these processes. This module must have access to read and edit both TCM system features and the CNC machine controller The module is expected to maintain the logic of the monitoring process to increase system accuracy. The cutoff from the CNC controller is expected to be dependent on the artificial intelligence classifiers that analyze the feedback data signals. In the case of logical classification, for example, if two successive segments are classified as anew and worn team, the module will diagnose this output depending on the system database, interrupt parameters, and tool state history. This output will be used to update the system baseline through the self- learning algorithm. The study presented for the detection of the tool prefault has shown the potential of acoustic emission (AE) signals to detect unstable crack advancement before tool wear or break in a period of 10 ms. More study is necessary to provide a more general system that can determine the stage of prefailure evolution, and that can quantitatively associate the AE signal with the amount of crack propagation for complex processing processes. At the same time, a reliable and robust tool failure and fault indication must be provided on time. The detection of AE waves related to the formation of new surfaces during unsteady crack propagation in intermittent cutting is difficult because of (1) the nonlinearity of the AE signal produced, (2) the nonstationary character of the stochastic unsteady crack propagation process, (3) the contamination of the crack propagation in the AE signal [9] , and (4) endless time bursts of high frequency bursts intrinsic in unsteady crack propagation. This is a reasonably short time on the millisecond ascend to take disciplinal measures after detective work. Besides, this requires the use of ultrahigh speed controllers to facilitate a fast stopping mechanism to overcome the inertia of stopping the cutting machine and the dominant online decision-making process in a brief time for HSM processes. Such issues will provide tremendous potential for increased process reliability, efficiency, and sustainability. The innovative aspect of the work by Tamang and Chandrasekaran is the optimization of the machinability parameters in the machining of the Inconel 825 aerospace alloy [10] . In conventional lathe tests, feed rate, spindle speed, and depth of cut parameters are defined as the input parameters, whereas the surface roughness is defined as the output parameter. The experimental study shows that |