Abstract:An efficient method based on Taguchi's design of experiment coupled with the grey relational analysis was studied, concentrating on the optimization of process parameters over surface roughness, cutting force and tool wear rate in milling of mild steel. This study consists of three stages: experimental work, single response optimization using Taguchi's S/N value and multi-response optimization using grey relational analysis. In the first stage, the experimental work was carried out using Taguchi's design of ex… Show more
“…The better surface roughness was obtained with minimum value of federate. 50,51 Figure 7.(a) Three-dimensional surface plots for interaction effects of cutting speed and feed rate, (b) cutting speed and depth of cut and (c) feed rate and depth of cut on surface roughness.…”
In this work, the flank wear of the cutting tool is predicted using artificial neural network based on the responses of cutting force and surface roughness. EN8 steel is chosen as a work piece material and turning test is conducted with various levels of speed, feed and depth of cut. Cutting force and surface roughness are measured for both the fresh and dull tool under dry cutting conditions. The tool insert used is CNMG 120408 grade, TiN coated cemented carbide tool. The experiments are conducted based on the response surface methodology face central composite design of experiments. The feed rate (14.52%), depth of cut (27.72%) and the interaction of feed rate and depth of cut (50.39%) influence the cutting force. The feed rate (21.33%) and the interaction of cutting speed and depth of cut (26.67%) influence the flank wear. The feed rate (61.63%) has the significant influence on surface roughness. The feed forward back propagation neural network of 5-n-1 architecture is trained using the algorithms like Levenberg Marquardt, BFGS quasi-Newton, and Gradient Descent with Momentum and Gradient descent with adaptive learning rate. The network performance has been assessed based on their mean square error and computation time. From this analysis, the BFGS quasi-Newton back propagation algorithm produced the least mean squared error value with minimum computation time.
“…The better surface roughness was obtained with minimum value of federate. 50,51 Figure 7.(a) Three-dimensional surface plots for interaction effects of cutting speed and feed rate, (b) cutting speed and depth of cut and (c) feed rate and depth of cut on surface roughness.…”
In this work, the flank wear of the cutting tool is predicted using artificial neural network based on the responses of cutting force and surface roughness. EN8 steel is chosen as a work piece material and turning test is conducted with various levels of speed, feed and depth of cut. Cutting force and surface roughness are measured for both the fresh and dull tool under dry cutting conditions. The tool insert used is CNMG 120408 grade, TiN coated cemented carbide tool. The experiments are conducted based on the response surface methodology face central composite design of experiments. The feed rate (14.52%), depth of cut (27.72%) and the interaction of feed rate and depth of cut (50.39%) influence the cutting force. The feed rate (21.33%) and the interaction of cutting speed and depth of cut (26.67%) influence the flank wear. The feed rate (61.63%) has the significant influence on surface roughness. The feed forward back propagation neural network of 5-n-1 architecture is trained using the algorithms like Levenberg Marquardt, BFGS quasi-Newton, and Gradient Descent with Momentum and Gradient descent with adaptive learning rate. The network performance has been assessed based on their mean square error and computation time. From this analysis, the BFGS quasi-Newton back propagation algorithm produced the least mean squared error value with minimum computation time.
“…In order to verify the dynamic test, designed dynamometer was undergone the actual machining test in L-Mill 55 vertical machining centre. The experiments were planned and carried out with mild steel based on Taguchi's L9 orthogonal array [44]. The experimental setup for measuring the cutting force was presented in Figure 10.…”
In this work, a milling dynamometer based on strain gauge with an octagonal and square ring was designed and tested. Strain gauges were attached with the mechanical rings to detect the deformation, during the machining process. Wheatstone bridge circuit was equipped with gauges to acquire the strain as voltage owing to the deformation of mechanical rings when machining takes place. The finite element analysis (FEA) was used to identify the location of maximum deformation and stress. The direction of rings and location of gauges were decided to increase the sensitivity and decrease the cross-sensitivity. Then, the cutting force was acquired through NI 6221 M series data acquisition (DAQ) card. The dynamometer had undergone a cycle of tests to verify its static and dynamic characteristics. The metrological characterization was performed according to the calibration procedure based on ISO 376 – 2011 standard. The cutting force was measured with both the dynamometers through milling experiments based on Taguchi’s L9 orthogonal array and the results were recorded. The measured cutting force varied from 300 N to 550 N. The obtained results depicted that low-cost milling dynamometer was reliable to measure the three component machining force. Overall, the square ring based dynamometer provides the better static and dynamic characteristics in terms of linearity, cross-sensitivity (4%), uncertainty (0.054%), and natural frequency (362.41 rev/s).
“…Rao and Murthy 15 used Response Surface Methodology (RSM), Artificial Neural network (ANN) and Support Vector Machines (SVM) to investigate the machinability characteristics of AISI 316 steel and concluded that cutting speed is the most influencing factor for machine vibration. Shankar et al 16 opined that best surface finished might be obtained in case minimum feed rate. Shankar et al 17 expressed that an increase in depth of cut increase the flank wear and cutting force simultaneously.…”
Surface finish is an important phenomenon in hard turning. There are many factors which can influence the finishing of a product. Literature review reveals that substantial research has been performed on hard machining, still relationship of tool wear and surface finish parameters like [Formula: see text] and [Formula: see text] is not established as the process is so dynamic and transient in nature. As a result, most of the responses like tool wear, surface integrity parameters, cutting force, and vibration are random in nature. In this investigation, Topic Modelling (TM), a relatively new topic particularly used in machine learning is applied to determine a particular stage of tool wear. Tool wear is divided into three distinct groups namely initial stage (IS), progressive stage (PS), and exponential stage (ES) from a number of experimental observations. Then, surface parameters namely [Formula: see text] and [Formula: see text] are measured. A probabilistic model consisting of tool wear and surface parameters is developed using Naïve based classifier. This model is capable to predict a particular stage of tool wear given randomly selected values of [Formula: see text] and [Formula: see text] To validate this probabilistic model, an alternative machine learning method called multinomial logistic regression is used. Each of this method indicates that the tool has reached to exponential stage when [Formula: see text] and [Formula: see text] =. [Formula: see text]
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
