PREDICTION OF REMAINING USEFUL LIFE OF AN END MILL CUTTER SEOW XIANG YUAN

Transcription

1 PREDICTION OF REMAINING USEFUL LIFE OF AN END MILL CUTTER SEOW XIANG YUAN Report submitted in partial fulfillment of the requirements for the award of the degree of Bachelor of Engineering (Hons.) in Manufacturing Engineering Faculty of Manufacturing Engineering UNIVERSITI MALAYSIA PAHANG JUNE 2016

2 iii SUPERVISOR S DECLARATION I hereby declare that I have checked this thesis and in my opinion, this thesis is adequate in terms of scope and quality for the award of the degree of Bachelor of Engineering in Manufacturing Signature : Name of supervisor Position : DR. MEBRAHITOM ASMELASH GEBREMARIAM : SENIOR LECTURER Date : 10/6/2016

3 iv STUDENT S DECLARATION I hereby declare that the work in this thesis is my own except for quotation and summaries which have been duly acknowledged. The thesis has not been accepted for any degree and is not concurrently submitted for award of other degree. Signature : Name ID Number : SEOW XIANG YUAN : FA12046 Date : 10/6/2016

4 viii TABLE OF CONTENTS Page SUPERVISOR S DECLARATION STUDENT S DECLARATION ACKNOWLEDGEMENTS ABSTRACT ABSTRAK TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES LIST OF SYMBOLS LIST OF ABBREVIATIONS iii iv v vi vii viii xii xiv xx xxii CHAPTER 1 INTRODUCTION 1.1 Introduction Problem Statement Objectives Project Scope 3 CHAPTER 2 LITERATURE REVIEW 2.1 Introduction Factors Affecting Tool Life 5

5 ix 2.3 Climb Milling and Conventional Milling Cutting Force in End Milling Calculations of FF Calculations of Fb Cutting Parameters in Milling Process Wear Mechanism of End Mill Remaining Useful Life (RUL) Tool Cutter Condition Monitoring Prediction Method Artificial Neural Network (ANN) Support Vector Regression (SVR) Summary 29 CHAPTER 3 DURABILITY ASSESSMENT METHODS 3.1 Introduction Flow Chart Project Descriptions Design of Experiment CAD Program Tool Wear Measurement Signal Acquisition and Processing Features Extraction and Reduction Support Vector Machine Regression Model Training Artificial Neural Network Model Training Predict for Remaining Useful Life Hardware and Software Application MAKINO Milling Machine HAAS Milling Machine Wire Cut Machine 67

6 x Metallurgical Microscope End Mill Insert Workpiece Material Dynamometer Amplifier Mastercam MATLAB 76 CHAPTER 4 RESULTS AND DISCUSSIONS 4.1 Introduction PHM Society Data Analysis Raw Data Features Extraction Features Reduction and Selection Support Vector Regression (SVR) Model Artificial Neural Network Model Remaining Useful Life (RUL) Prediction Performance Assessment Actual Tool Wear Measurement Raw Data Features Extraction Features Reduction and Selection Support Vector Regression (SVR) Model Artificial Neural Network Model Remaining Useful Life (RUL) Prediction Performance Assessment 108 CHAPTER 5 CONCLUSION AND RECOMMENDATIONS 5.1 Introduction Conclusion 112

7 xi 5.3 Recommendations 113 REFERENCES 114 APPENDICES 119 A Budget Plan 119 B Gantt Chart 120 C CAD Program 122

8 xii LIST OF TABLES Table No. Title Page 2.1 ANOVA for surface roughness ANOVA for surface roughness Important statistical features Experiment setting Wire cut machine specification Cutting Condition Significant statistical features Stepwise regression parameters Features selection result from stepwise regression Support vector machine regression model Artificial Neural network setting Mean Square Error and regression of the model Performance assessment for SVM Regression model Performance assessment for neural network model Important statistical features Stepwise regression setting Features selection result from stepwise regression 97

9 xiii 4.13 Support vector machine regression model Neural network setting Mean Square Error and regression of the model Performance assessment for models trained 110

10 xiv LIST OF FIGURES Figure No. Title Page 2.1 Conventional Milling and Climb milling Milling process Graph for cutting force Graph for surface roughness Central wear and flank wear Schematic illustration of tool wear Chipping and catastrophic fracture on end mill Concept illustration of remaining life of an asset Remaining useful life (RUL) Prognostics approach ANN architecture ANN output graph Health indicator result using EM-PCA (left) and ISOMAP (right) Result of SVR model Flow chart Raw material 34

11 xv 3.3 Wire cutting of material Material surface after wire cut Surface finish from face milling process Experiment setup Edge finding process Cutting process Tool holder and insert Mastercam Mill X5 software Stock dimension setup Tool parameters Linking parameters Tool path design Tool path illustration Tool wear measurement Tool wear measurement process Computer and microscope system Light level on microscope Specimen preparation Microscope software 47

12 xvi 3.22 Table height adjustment Image capture Wear measurement Data acquisition and export process Dynoware Amplifier Connection Measuring time and sampling rate Sensor data measurement Data exportation Exported data Features extraction Features reduction Support Vector Machine regression model training Predictors and respond fitrsvm function code predict function code Artificial Neural Network model training Neural fitting app Data loading 59

13 xvii 3.41 Validation and test data Neurons selection and network architecture Training of network Retraining interface Network result saving Predicting remaining useful life Makino KE55 CNC milling machine Specifications for Makino KE HAAS milling machine VF Control unit Sodick VZ 300L wire cut machine Olympus BX51M metallurgical microscope End mill insert Insert specification Workpiece physical data Dynamometer 9257B Dynamometer technical data Amplifier type 5070A Amplifier specifications 75

14 xviii 3.60 Mastercam logo MATLAB Logo Raw data from force sensor during first cutting process (left) 80 and final cutting process (right) 4.2 Raw data from vibration sensor during first cutting process (left) 80 and final cutting process (right) 4.3 Raw data from acoustic sensor during first cutting process (left) 81 and final cutting process (right) 4.4 Comparison between actual wear and predicted wear of SVR Neural network architecture Mean Square Error Regression plot Comparison between actual wear and predicted wear of neural network Comparison between actual RUL and neural network RUL Comparison between actual RUL and Support Vector Regression RUL Actual tool wear of end mill Tool wear after first cutting process Tool wear after last cutting process Comparison between actual wear and predicted wear of SVR Neural network architecture Mean Square Error 104

15 xix 4.17 Regression plot Comparison between actual wear and predicted wear of neural network Illustration for remaining useful life calculation Comparison between actual RUL and Support Vector Regression RUL Comparison between actual RUL and Artificial Neural Network RUL 108

16 xx LIST OF SYMBOLS C Force Coefficient C Penalty function Ɛ Epsilon Ɛ(t) Difference between predicted and actual RUL ϵ Mean of total difference between predicted and actual RUL e Exponent F Total cutting force FB Bottom force FF Flank force h Instantaneous uncut chip thickness N Number of cuts R 2 R squared RULactual Actual remaining useful life RULpredicted Predicted remaining useful life t Current time tf Final time V Voltage W Axial length

17 xxi Summation ⱷ Angle

18 xxii LIST OF ABBREVIATIONS AE Acoustic emission AISI American Iron & Steel Institute ANN Artificial Neural Network ANOVA Analysis of Variance CART Classification and Regression Tree CNC Computer Numerical Control CMS Condition Monitoring System DAQ Data Acquisition EM-PCA Expected Maximization Principal Component Analysis ESR Electro-Slag-Refining GUI Graphical User Interface HP Horse Power HSS High Speed Steel MAPE Mean Absolute Percentage Error MSE Mean Square Error PHM Prognostics and Health Management PVD Physical Vapor Deposition QP Quadratic Programming

19 xxiii RMS Root Mean Square RUL Remaining Useful Life SMO Sequential Minimal Optimization SVR Support Vector Regression TCM Tool Condition Monitoring WPD Wavelet Packet Decomposition