Prediction of Vibrations & Cutting force of single point cutting tool in turning by using Artificial Neural Network


Professorial Dissertation, 2015

95 Pages


Excerpt


CONTENTS

Acknowledgement

Abstract

List of Figures

List of Tables

Nomenclature

1 Introduction
1.1 Introduction to Machine Vibration
1.2 Neural Network
1.3 Artificial Neural Network
1.4 Mathematical model of Artificial Neural Network
1.6 Objectives of the Present Work
1.7 Methodology
1.8 Organization of the Report
1.9 Summary

2 Literature Review
2.1 Introduction
2.2 Artificial Neural Network
2.3 Cutting tool vibrations
2.4 Cutting force
2.5 Summary

3 Analysis of vibrations & forces in turning
3.1 Introduction
3.2 Vibrations on single point cutting tool in cutting
3.2.1 Types of vibrations
3.2.2 Vibration measurement using FFT Analyzer
3.3 Forces on single point cutting tool in cutting
3.4 Summary

4 Experimentations
4.1 Introduction
4.2 Cutting Material
4.3 Cutting tool Material
4.3.1 Carbides
4.4 Adjustable Cutting Parameters in Turning as Input
4.4.1 Cutting speed
4.4.2 Feed
4.4.3 Depth of cut
4.5 Output parameters in Turning
4.5.1 Tool vibrations
4.5.2 Cutting Forces acting on single point cutting tool
4.6 Instrumentations
4.6.1 Lathe machine
4.6.2 Digital Tachometer
4.6.3 FFT Analyzer
4.6.4 PC
4.6.5 Tool Dynamometer
4.7 Experimental setup & Instrumentations
4.8 Summary

5 Programming Artificial Neural Network in Matlab
5.1 Introduction
5.2 Creating Artificial Neural Network
5.2.1 Steps to create the ANN models
5.2.1.1 Data collection
5.2.1.2 Building the network
5.2.1.3 Training the network
5.2.1.4 Testing the network
5.2.1.5 Simulate the network
5.3 Neural Network Graphical User Interface in Matlab
5.3.1 Create Input & Target files in Matlab
5.3.2 Set the training data
5.3.3 To begin using the NN GUI
5.3.4 Click on Import to import data
5.3.5 Click on New to create your neural network
5.3.6 Training the network
5.3.7 Click on Simulate tab in Network window
5.3.8 Exporting Errors/Output to Matlab Workspace
5.4 Summary

6 Results and Discussions
6.1 ANN model for prediction of tool vibrations
6.2 ANN model for prediction of cutting force

7 Conclusions and Future Scope

7.1 Conclusion

References

Acknowledgement

While working on the current dissertation is just not an individual contribution till its completion. I take this opportunity to thank all for bringing it close to the conclusion.

First of all, I would like to thank Prof. L. B. Raut for guide and continuously assessing my work providing great guidance by timely suggestions and discussions at every stage of this work.

I am thankful to Dr. P. S. Kachare, Head of Department of Mechanical engineering for providing all facilities, without which this dissertation work would not have been possible. I sincerely thank to Dr. B.P. Ronge, Principal SVERI’s COE Pandharpur.

I thank my Father, Mother, Brother and last but not least my wife, also my friends and my all relatives for their persistent support during the studies. Also I must acknowledge all those who helped me directly and indirectly for completing my project.

Mr. Matin A. Shaikh

Abstract

The objective of this work is to develop a model to simulate the vibrational effects of rotating machine parts on the single point cutting tool. In this experimental studies were performed on turning process. In this study tool vibrations & cutting force acting on the single point cutting tool is to be calculated i.e. the tool vibrations & cutting force are the outputs of this study. And input parameters are the spindle speed, feed & depth of cut. By varying these three input parameters we can get two output parameters. The tool vibration is measured with the help of accelerometer along with a device called as Fast Fourier Transformer (FFT) Analyzer.

The vibration of single point cutting tool is sensed by accelerometer located on the tool-post of lathe machine. The accelerometer will send the sensed vibration to FFT Analyzer which can be convert the sensed data by using accelerometer shown in PC such as frequency, Amplitude, displacement & so on. This turning process is orthogonal machining process is considered therefore in this study the forces acting on cutting tool are only two components are obtained. The cutting force is to be measured with the help of strain gauge type tool dynamometer. The tool dynamometer consists of strain gauge compacted tool post & control unit. The tool post is mounted on the lathe machine & in that tool post the cutting tool is attached. The sensed two components of cutting force is send by the strain gauges to its control unit & that control unit displays the respective force.

The two components of cutting forces has resultant is to be calculated & the tool vibrations & net cutting force are the obtained by experimentations. The turning process is dry machining process because the cutting tool is attached to the strain gauge compacted tool post & these strain gauges are connected to its control unit & also to measure the tool vibrations, the accelerometers is mounted on that tool post, So that at the cutting tool very sensational electronic devices are present so dry machining is to be considered. The obtained experimental data given to an Artificial Neural Network (ANN) in Matlab, with the help of experimental data ANN is to be trained, and by using ANN can predict the vibrations by changing parameters of turning such as spindle speed, feed & depth of cut.

This model of ANN can be predict vibrations of single point cutting tool to avoid the failure of cutting tool. Then there is no need of future experiment to calculate the tool vibrations & cutting force acting on the single point cutting tool in turning process. With the help of this trained artificial neural network model it can predict the tool vibrations & cutting force of any three machining parameters like spindle speed, feed & depth of cut.

Keywords: Vibration, Cutting force, Orthogonal cutting, ANN.

List of Figures

Abbildung in dieser Leseprobe nicht enthalten

List of Tables

Abbildung in dieser Leseprobe nicht enthalten

Nomenclature

Abbildung in dieser Leseprobe nicht enthalten

CHAPTER 1 Introduction:

1.1 Introduction to Machine vibration

Much emphasis has been placed upon vibrations in machine tools during recent years because many people have recognized that accuracy, surface finish and, last but not least, production costs are considerably influenced by them. Today a collection of sophisticated instruments is available for the investigation of machine tool vibration. However, in the final analysis, the finished surface itself will reflect the dynamic behavior of the machine tool.

Cutting tool have always vibrated and will continue to do so. It strives to measure these vibrations and keep it at or below a tolerable level. This was easier to do in the past than it is today. While higher cutting speeds generally contribute to an improvement of the surface finish obtained, they often excite components of the machine tool at their natural frequency.

Machine operation confronted by a shortage of technical manpower and pricing competition not only need to implemented automated and operator free technology, but also needed to meet requirement of precision through process planning thus achieving maximum productivity, meaning a cutting condition with an optimal metal removal rate. In metal cutting operation one of the major obstacles to realizing its complete automation is that of cutting tool -state prediction, where the tool wear, cutting force and vibration of cutting tool is important factor in productivity and efficiency of manufacturing.

The exciting force is trying to cause vibration, whereas the stiffness, mass and damping forces are trying to oppose the exciting force and control or minimize the vibration. Perhaps the simplest and easiest way to demonstrate and explain vibration and its measurable characteristics is to follow the motion of a weight suspended by a spring. This is a valid analogy since all machines and their components have weight (mass), spring-like properties (stiffness) and damping. The motion of the mass from top to bottom range and back to the initial starting position in the vertical direction is referred to as one cycle, and it has all the characteristics needed to define the vibration. Continued motion of the spring-mass system will simply be repeating these measurable characteristics.

In metal cutting operation our goals to increasing productivity, reliability and quality of work piece through prediction of vibration of cutting tool using some developed neural network. In fact the difference between real and theoretical, surface quality can be attributed to the influence of physical and dynamic phenomena such as tool vibration, friction of cut surface against tool point, cutting force.

The amplitude and frequency of cutting force, torque and power are used in sizing machine tool structures, spindle and feed drive mechanisms, bearings, motors, drives as well as the shank size of the tool and fixture rigidity. The stress and temperature field in the cutting tool edge, chip and work piece surface are used in designing cutting edge shape as well as feed, depth of cut, to avoid residual stresses on the finished surface.

Continuous improvement in the technological performance of machining operation have been sought through research and development including new or more wear resistant tool material as well as new geometrical design. A little has been reported on the quantitative assessment and information of hard surface coating in terms of cutting force to guide the selection and design of machine tool cutting tools and selection of economic cutting condition.

A Vibration analysis has revealed that dynamic force, related to cutting condition, amplitude of tool vibration at resonance and variation of tool’s natural frequency, methods for co-relating measured process parameters fail primarily in to three categories. First categories knowledge only of process parameter such as mathematical modeling, second categories knowledge of the process and grouped together such as FFT Analysis and last category knowledge of based systems such as neural network. A neural network build upon the modeling technique are able to represent complex and uncertain relationship between input and output variables.

1.2 Neural Network

A neuron operates by receiving signals from other neurons through connections, called synapses. The combination of these signals, in excess of a certain threshold or activation level, will result in the neuron firing that is sending a signal on to other neurons connected to it. Some signals act as excitations and others as inhibitions to a neuron firing. The thinking is believed to be the collective effect of the presence or absence of firings in the pattern of synaptic connections between neurons.

This sounds very simplistic until we recognize that there are approximately one hundred billion neurons each connected to as many as one thousand others in the human brain. The massive number of neurons and the complexity of their interconnections results in a "thinking machine", i.e. human brain.

Each neuron has a body, called the soma. The soma is much like the body of any other cell. It contains the cell nucleus, various bio-chemical factories and other components that support ongoing activity.

Surrounding the soma are dendrites. The dendrites are receptors for signals generated by other neurons. These signals may be excitatory or inhibitory. All signals present at the dendrites of a neuron are combined and the result will determine whether or not that neuron will fire.

If a neuron fires, an electrical impulse is generated. This impulse starts at the base, called the hillock, of a long cellular extension, called the axon, and proceeds down the axon to its ends.

Prediction of Vibration & Cutting force of single point cutting tool in Turning by using Artificial Neural Network

Fig. No. 1.1 Structure of Neuron

Abbildung in dieser Leseprobe nicht enthalten

The end of the axon is actually split into multiple ends, called the boutons. The boutons are connected to the dendrites of other neurons and the resulting interconnections are the previously discussed synapses. (Actually, the boutons do not touch the dendrites; there is a small gap between them.) If a neuron has fired, the electrical impulse that has been generated stimulates the boutons and results in electrochemical activity which transmits the signal across the synapses to the receiving dendrites.

At rest, the neuron maintains an electrical potential of about 40-60 mill volts. When a neuron fires, an electrical impulse is created this is the result of a change in potential to about 90-100 mill volts. This impulse travels between 0.5 to 100 meters per second and lasts for about 1 millisecond. Once a neuron fires, it must rest for several milliseconds before it can fire again. In some circumstances, the repetition rate may be as fast as 100 times per second, equivalent to 10 milliseconds per firing.

Each neuron receives electrochemical inputs from other neurons at the dendrites. If the sum of these electrical inputs is sufficiently powerful to activate the neuron, it transmits an electrochemical signal along the axon, and passes this signal to the other neurons whose dendrites are attached at any of the axon terminals. These attached neurons may then fire.

It is important to note that a neuron fires only if the total signal received at the cell body exceeds a certain level. The neuron either fires or it doesn’t, there aren’t different grades of firing.

So, our entire brain is composed of these interconnected electro-chemical transmitting neurons. From a very large number of extremely simple processing units (each performing a weighted sum of its inputs, and then firing a binary signal if the total input exceeds a certain level) the brain manages to perform extremely complex tasks. This is the model on which artificial neural networks are based.

Thus far, artificial neural networks haven’t even come close to modeling the complexity of the brain, but they have shown to be good at problems which are easy for a human but difficult for a traditional computer, such as image recognition and predictions based on past knowledge.

Compare this to a very fast electronic computer whose signals travel at about 200,000,000 meters per second (speed of light in a wire is 2/3 of that in free air), whose impulses last for 10 nanoseconds and may repeat such an impulse immediately in each succeeding 10 nanoseconds continuously. Electronic computers have at least a 2,000,000 times advantage in signal transmission speed and 1,000,000 times advantage in signal repetition rate.

It is clear that if signal speed or rate were the sole criteria for processing performance, electronic computers would win hands down. What the human brain lacks in these, it makes up in numbers of elements and interconnection complexity between those elements. This difference in structure manifests itself in at least one important way; the human brain is not as quick as an electronic computer at arithmetic, but it is many times faster and hugely more capable at recognition of patterns and perception of relationships.

The human brain differs in another, extremely important, respect beyond speed; it is capable of “self-programming” or adaptation in response to changing external stimuli. In other words, it can learn. The brain has developed ways for

Prediction of Vibration & Cutting force of single point cutting tool in Turning by using Artificial Neural Network

neurons to change their response to new stimulus patterns so that similar events may affect future responses.

Abbildung in dieser Leseprobe nicht enthalten

Fig. No. 1.2 Neurons connection

Abbildung in dieser Leseprobe nicht enthalten

Fig. No. 1.3 Neuron with synapse

Abbildung in dieser Leseprobe nicht enthalten

Fig. No. 1.4 A Biological model of Neuron

The perceptron is a mathematical model of a biological neuron. While in actual neurons the dendrite receives electrical signals from the axons of other neurons, in the perceptron these electrical signals are represented as numerical values. At the synapses between the dendrite and axons, electrical signals are modulated in various amounts. This is also modeled in the perceptron by multiplying each input value by a value called the weight. An actual neuron fires an output signal only when the total strength of the input signals exceeds a certain threshold. The model in this phenomenon in a perceptron by calculating the weighted sum of the inputs to represent the total strength of the input signals, and applying a step function on the sum to determine its output. As in biological neural networks, this output is fed to other perceptron.

ANNs are processing devices (algorithms or actual hardware) that are loosely modeled after the neuronal structure of the all animal cerebral cortex but on much smaller scales. A large ANN might have hundreds or thousands of processor units, whereas a mammalian brain has billions of neurons with a corresponding increase in magnitude of their overall interaction and emergent behavior.

[...]

Excerpt out of 95 pages

Details

Title
Prediction of Vibrations & Cutting force of single point cutting tool in turning by using Artificial Neural Network
Author
Year
2015
Pages
95
Catalog Number
V315987
ISBN (eBook)
9783668159778
ISBN (Book)
9783668159785
File size
4070 KB
Language
English
Keywords
Vibration, Cutting tool, Turning, ANN, Prediction, Cutting force, Orthogonal cutting
Quote paper
Matin Shaikh (Author), 2015, Prediction of Vibrations & Cutting force of single point cutting tool in turning by using Artificial Neural Network, Munich, GRIN Verlag, https://www.grin.com/document/315987

Comments

  • No comments yet.
Look inside the ebook
Title: Prediction of Vibrations & Cutting force of single point cutting tool in turning by using Artificial Neural Network



Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free