Uday S. Dixit

Prediction of surface roughness and dimensional deviation by measuring cutting forces and vibrations in turning process

Abstract Prediction of surface finish and dimensional deviation is an essential prerequisite for developing an unmanned turning center. It is also important for optimization of turning process. In this work, it is found that, using neural network, surface finish can be predicted within a reasonable degree of accuracy by taking the acceleration of radial vibration of tool holder as a feedback. It is also possible to utilize the fitted network for predicting the surface finish in turning with a tool of same material but different geometry provided coolant situation is same. For that purpose, only few experiments are needed with the new tool to modify the neural network predicted results. However, different neural network models have to be fitted for dry and wet turning, as well as for turning by HSS and carbide tools. It was observed that while turning the steel rod with TiN coated carbide tool, surface finish improves with increasing feed up to some feed where from it starts deteriorating with further increase of feed. This type of behavior is not observed in turning with HSS tool. Dimensional deviation is significant only in the case of turning of a slender work-piece. Hence, neural network prediction models are developed separately for that. Radial component of cutting force and acceleration of radial vibration were taken as a feedback to predict dimensional deviation. The performance of the developed neural network models is assessed by carrying out a number of experiments involving dry and wet turning of mild steel rods using HSS and carbide tools.

Application of neural networks in generating processing map for hot working

Abstract An important parameter in the mechanical working of materials is called workability, which is the relative ease with which a metal can be shaped through plastic deformation without the formation of any defect. Workability can be evaluated by means of processing maps, constructed from experimentally generated flow stress variation with respect to strain, strain rate and temperature. The present work demonstrates the use of neural network in generating processing maps for hot working processes. A neural network model was trained and tested for predicting the flow stress by taking data available in the literature for 99.99% pure aluminum. It was found that the trained neural network could predict the flow stress for unseen data quite reliably. At strain of 0.4, power dissipation and instability maps were constructed, utilizing the flow stress prediction by neural network. Superimposition of these maps provided processing maps at 0.4 strain, which was similar to that available in the literature. This established the potential of applying neural network, which is more robust technique than conventional method, for generating the processing map.

Environmentally Friendly Machining

Machining is a controlled material removal process and finds its application in a variety of industrial sectors such as automobile, aerospace, and defense. Similar to many other manufacturing processes, machining bears significant environmental impacts in terms of energy/resource consumption, airborne emissions, wastewater discharge, and solid wastes along with occupational health risks. Most of these issues are due to the use of cutting fluids, which are traditionally formulated with petroleum-derived compounds with high ecotoxicity and low biodegradability. Exposure to these chemicals, along with growth of microorganisms and biocides used for microbial control, could lead to respiratory irritation, asthma, pneumonia, dermatitis, and even cancer. To address these concerns, extensive effort has been put forth to (1) extend the cutting fluid life span by removing particulates, free oils, and other contaminants via separation and filtration, (2) reformulate traditional petroleum-based fluids with vegetable oils and bio-based ingredients for lower toxicity and higher biodegradability, and (3) reduce or even eliminate the reliance on cutting fluids during machining through dry machining and minimum quantity lubrication (MQL) techniques. Apart from these technology developments, machining process parameters can be optimized for reduced environmental impacts, especially energy consumption and carbon footprint. Process optimization approaches require the development of models and equations to correlate process parameters with process inputs and outputs. Given the current status in the field, opportunities exist in designing new bio-based, microfiltration-compatible formulations using industrial by-products, optimizing minimum MQL system configuration, advancing cutting tool insert materials and lubricants for MQL, and developing high-energy efficiency machine tools.

An analysis of the steady-state wire drawing of strain-hardening materials

Abstract A comprehensive investigation of the steady-state wire drawing process has been done to study the effects of various process variables on important drawing parameters and deformation, the process variables considered being the reduction ratio, the die semi-angle, the coefficient of friction and the back tension, whilst the drawing parameters studied are the die-pressure, the drawing stress and the separation force. The deformation is represented by contours of equivalent strain and of equivalent strain-rate. The quantitative effects of strain hardening on the drawing parameters and qualitative effects on the deformation are studied also. A comparison of the drawing parameters is made for three materials (copper, aluminium and steel).

Application of radial basis function neural networks in optimization of hard turning of AISI D2 cold-worked tool steel with a ceramic tool

Abstract In this work, the optimization of a finish hard turning process for the machining of D2 steel with ceramic tools is carried out. With the help of replicate experimental data at 27 different cutting conditions, radial basis function neural network models are fitted for predicting the surface roughness and tool wear as functions of cutting speed, feed, and machining time. A novel method for neural network training is proposed. The trained neural network models are used as a black box in the optimization routine. Two types of optimization goal are considered in this work: minimization of production time and minimization of the cost of machining. One novel feature of this work is that the surface roughness is considered in the tool life instead of as a constraint. This is possible owing to the availability of the relationship of surface roughness with time in the neural network model. The results of optimization will be dependent on the tool change time and the ratio of operating cost to tool change cost. The results have been presented for the possible ranges of these parameters. This will help to choose the appropriate process parameters for different situations, and a sensitivity analysis can be easily carried out.

Micromanufacturing: A review—part II

Micromanufacturing processes are expanding in their length and breadth as long as the related research and development (R&D) activities and applications are concerned. Products are getting miniaturized and their performance efficiency is getting enhanced by the addition of micro/nanofeatures and devices. In the set of these two articles (Part I and Part II), an attempt has been made to review the latest R&D activities of the selected micromanufacturing processes. This article (Part I) deals with a review of the literature related to attrition (subtractive, or machining and finishing) processes (both types—traditional and advanced) including microturning, micromilling, microdrilling, abrasive jet micromachining, laser beam micromachining, electrochemical micromachining, magnetorheological finishing, abrasive flow finishing, magnetic abrasive finishing, ion beam micromachining and so on. Apart from the subtractive processes, an overview of the X-ray lithography has also been presented. An attempt has been made to report some applications to help the readers to evolve more new applications of these processes. At the end of different sections/subsections, some research areas have been identified, which would hopefully fill the gaps between the theoretical analysis, experimental work and applications.

A neural network-assisted finite element analysis of cold flat rolling

The finite element analysis of the cold flat rolling process is well established. However, the requirement of large computational time makes it unsuitable for online applications. Recently, there have been some applications of modeling the rolling process by means of neural networks. In most of the previous works, trained networks predict only roll force and roll torque. The input data for training the neural network have been obtained either through experiments or from finite element method (FEM) code. In this work, the neural networks have been used for predicting the velocity field and location of neutral point. The training data are obtained from a rigid-plastic finite element code. The trained network provides a suitable guess for the velocity field and location of the neutral point, that is further refined by the finite element code. The post-processor of the FEM code computes roll force, roll torque, strain distribution, etc. This procedure provides highly accurate solution with reduced computational time and is suitable for on line control or optimization.

A finite element analysis of flat rolling and application of fuzzy set theory

In this work, a model for steady-state plane strain cold rolling of a strain hardening material is proposed. The mixed pressure and velocity formulation is used and front and back tensions are included in the model. Roll deformation is taken into account by Hitchcocks formula and the friction model of Wanheim and Bay is used. Comparisons with the experimental results found in the literature are made to evaluate the accuracy of the present model. In the rolling process, material properties and friction coefficients are not known precisely and hence they can be treated as fuzzy numbers. Analysis with the fuzzy parameters is carried out to highlight the usefulness of such an analysis. A method to assess the reliability of a design is also proposed.

A study on residual stresses in rolling

Determination of residual stresses in rolled material is important for the design of product and process. Different approaches for determining them and difficulties associated with the methods are discussed. A simplified approach to find the longitudinal residual stress (stress in the direction of rolling) is proposed. A parametric study has been carried out to show the influence of process parameters on the residual stress distribution pattern.

Shape Optimization of Flexible Robotic Manipulators

In this work, the problem of shape optimization of flexible robotic manipulators of circular cross sections is studied. Two different manipulators are considered-a manipulator with revolute joint and a roller supported Cartesian manipulator. The finite element method is used to find the natural frequency and dynamic response of a flexible manipulator by treating it as an Euler-Bernoulli beam. The cross-sectional diameter is varied along the length keeping the constraint on the mass of the manipulator and static tip deflection in order to maximize the fundamental frequency of the beam. This optimization problem is compared with other optimization problems (with different objective functions and constraints). It is observed that the proposed optimization problem is superior to other optimization problems.