Data Driven Smart Manufacturing Technologies and Applications (Springer Series in Advanced Manufacturing)

✍ Scribed by Weidong Li (editor), Yuchen Liang (editor), Sheng Wang (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 224
• Category: Library

✦ Synopsis

This book reports innovative deep learning and big data analytics technologies for smart manufacturing applications. In this book, theoretical foundations, as well as the state-of-the-art and practical implementations for the relevant technologies, are covered. This book details the relevant applied research conducted by the authors in some important manufacturing applications, including intelligent prognosis on manufacturing processes, sustainable manufacturing and human-robot cooperation. Industrial case studies included in this book illustrate the design details of the algorithms and methodologies for the applications, in a bid to provide useful references to readers.

Smart manufacturing aims to take advantage of advanced information and artificial intelligent technologies to enable flexibility in physical manufacturing processes to address increasingly dynamic markets. In recent years, the development of innovative deep learning and big data analytics algorithms is dramatic. Meanwhile, the algorithms and technologies have been widely applied to facilitate various manufacturing applications. It is essential to make a timely update on this subject considering its importance and rapid progress.

This book offers a valuable resource for researchers in the smart manufacturing communities, as well as practicing engineers and decision makers in industry and all those interested in smart manufacturing and Industry 4.0.

✦ Table of Contents

Preface
Contents
About the Editors
Introduction
1 Smart Manufacturing Trend
2 Data Driven Smart Manufacturing
3 Typical Deep Learning Models
3.1 Some Development of Deep Learning Models
3.2 Typical Deep Learning Algorithms
3.3 Further Discussions on Data Driven Smart Manufacturing
References
Fog Computing and Convolutional Neural Network Enabled Prognosis for Machining Process Optimization
1 Introduction
2 Literature Survey
3 System Framework and Workflow
4 Detailed Design and Algorithms
4.1 Terminal Layer
4.2 Fog Layer
4.3 Cloud Layer
4.4 Time Complexity Analysis
5 System Deployment and Analysis
5.1 System Setup and Deployment
5.2 Analysis on Gaussian Kernel Modelling and Fault Index
5.3 Analysis on Different Structures of the CNN Design
5.4 Analysis on System Performance Improvement
5.5 Analysis on Improved Efficiency of Machining System
6 Conclusions
References:
Big Data Enabled Intelligent Immune System for Energy Efficient Manufacturing Management
1 Introduction
2 Literature Survey
3 System Framework
4 Intelligent Immune Mechanism
4.1 Processing of Monitored Energy Data
4.2 The Immune Mechanism
4.3 Re-Scheduling Optimization
5 Industrial Deployment and Case Studies
5.1 Set-Up of Machining Processes and Energy Data Collection
5.2 Big Data Partition and ANNs Training
5.3 Immune Processes
6 Conclusions
References
Adaptive Diagnostics on Machining Processes Enabled by Transfer Learning
1 Introduction
2 Literature Survey
2.1 Related Work on Fault Diagnostics
2.2 Related Work on Transfer Learning Based Applications
3 System Framework
3.1 Manufacturing System and Machining Process Lifecycle
3.2 Flow of Adaptive Diagnostics
4 CNN Designed for Diagnostics on Machining Process
5 Adaptive Diagnostics on Machining Process Lifecycle
5.1 Forward Computing Using Optimally Aligned Data
5.2 Back Propagation Optimization for Minimizing Feature Distribution Discrepancy and Classification Error
6 Experiments and Discussions
6.1 Experiment Set-Up
6.2 Optimized Alignment of the Datasets Between Domains
6.3 Minimization of Feature Distribution Discrepancy Between Domains and Classification Error
7 Conclusions
References
CNN-LSTM Enabled Prediction of Remaining Useful Life of Cutting Tool
1 Introduction
2 Literature Survey
3 Methodology and System Flow
3.1 Signal Partition Based on the Hurst Exponent
3.2 A Hybrid CNN-LSTM Algorithm for Prediction
4 Case Study and Methodology Validation
4.1 Experimental Setup
4.2 Signal Partition Based on the Hurst Exponent
4.3 Performance Evaluation on the Hurst Exponent and CNN-LSTM
4.4 Comparison of the Hurst Exponent with Other Methods
4.5 RUL Prediction
5 Conclusions
References:
Thermal Error Prediction for Heavy-Duty CNC Machines Enabled by Long Short-Term Memory Networks and Fog-Cloud Architecture
1 Introduction
2 Related Works
2.1 Data-Based Modelling for Thermal Error Prediction
2.2 STM Networks for Data-Based Manufacturing Applications
2.3 Information Architecture for Data-Based Modelling
3 Research Methodologies and System Design
3.1 FEA Analysis for Optimized Deployment of Sensors
3.2 Improved Grey Relational Analysis (iGRA)
3.3 LSTM Design for Thermal Error Prediction
3.4 Fog-Cloud Architecture Design
4 Case Studies and Experimental Analysis
4.1 Sensor Deployment
4.2 Analysis on iGRA
4.3 Analysis on Training the LSTM Networks
5 Conclusions
References
Deep Transfer Learning Enabled Estimation of Health State of Cutting Tools
1 Introduction
2 Related Works
2.1 Deep Learning for PHM
2.2 Transfer Learning Enabled Deep Learning
2.3 Images in PHM Applications
3 Methodology
3.1 Problem Definition and Overall Methodology
3.2 The Input Data for Transfer Learning
3.3 CNNs Models and Regression Stack
3.4 Transfer Learning Process
4 Experimental Results and Discussions
4.1 Experiment Data Collection
4.2 Result Analysis and Observations
5 Conclusions
References
Prediction of Strength of Adhesive Bonded Joints Based on Machine Learning Algorithm and Finite Element Analysis
1 Introduction
2 Literature Survey
3 System Structure
4 Detailed System and Algorithm
4.1 Experiment Stage
4.2 FEA Model Stage
4.3 Machine Learning Model and Optimization Stage
5 System Development and Analysis
5.1 Experiment and FEM Validation
5.2 Machine Learning Model
5.3 Optimization
6 Conclusions
References
Enhancing Ant Colony Optimization by Adaptive Gradient Descent
1 Introduction
2 Background
2.1 Ant Colony Optimization for TSP
2.2 Stochastic Gradient Descent in DL
3 Adaptive SGD-Based Ant Colony Optimization
3.1 Loss Function and Gradient Definition for ACO
3.2 Calculation of Gradient in ACO
3.3 Design and Implementation of ADACO
3.4 Complexity
4 Performance Evaluation
4.1 Convergence Analysis
4.2 Parameter Adaptation
4.3 Solution Quality
4.4 Comparison with Other High Performance TSP Solvers
4.5 Time Cost
5 Related Works
5.1 Improving ACO Algorithms
5.2 Models to Solving TSPs
6 Conclusions
References
Index