Digital Twins of Advanced Materials Processing

Chapter 1: Introduction

• • 1.1 An inspiring event
• • 1.2 What is a digital twin?
• • 1.3 Evolution and history of digital twins
• • 1.4 Capabilities and uses of digital twins in materials processing industries
• • 1.5 Solution of major challenges in materials processing
• • 1.6 Content of this book

Chapter 2: Building blocks of a digital twin

• • 2.1 Introduction
• • 2.2 Mechanistic models
• • 2.3 Machine learning and deep learning
• • 2.4 Surrogate and reduced order models
• • 2.5 Statistical models
• • 2.6 Control models
• • 2.7 Big data and data analytics
• • 2.8 Interconnection among the building blocks

Chapter 3: Mechanistic models

• • 3.1 Introduction
• • 3.2 Mechanistic models of manufacturing processes
• • 3.2.1 Heat transfer and fluid flow modeling
• • 3.2.2 Process modeling
• • 3.3 Mechanistic modeling of microstructure
• • 3.3.1 Solidification morphology
• • 3.3.2 Grain structure
• • 3.3.3 Microstructure evolution
• • 3.4 Mechanistic modeling of mechanical properties
• • 3.5 Mechanistic modeling of performance
• • 3.5.1 Defects
• • 3.5.2 Mitigating composition change - selective vaporization of alloying elements
• • 3.5.3 Modeling residual stress and distortion
• • 3.6 Software for mechanistic modeling of manufacturing

Chapter 4: Surrogate and reduced order models

• • 4.1 Introduction
• • 4.2 Analytical models
• • 4.3 Dimensionless number-based calculations
• • 4.4 Reverse models
• • 4.5 Back of the envelop calculations
• • 4.6 Data-driven surrogate models
• • 4.6.1 Linear regression
• • 4.6.2 Support vector regression
• • 4.6.3 Radial basis functions
• • 4.6.4 Kriging
• • 4.6.5. Mixture of surrogates
• • 4.6.6 Available software

Chapter 5: Machine learning and deep learning

• • 5.1 Introduction
• • 5.2 Machine learning algorithms and applications
• • 5.2.1 Regression algorithms
• • 5.2.2 Classification algorithms
• • 5.3 Deep learning algorithms and applications
• • 5.3.1 Discriminative deep learning
• • 5.3.2 Generative deep learning
• • 5.3.3 Reinforcement learning
• • 5.4 Image processing and feature extraction
• • 5.5 Open-source packages
• • 5.6 Data needs

Chapter 6: Statistical models

• • 6.1 Introduction
• • 6.2 Different statistical models for digital twins
• • 6.2.1 Regression Models
• • 6.2.2 Time Series Analysis
• • 6.2.3 Monte Carlo Simulations
• • 6.2.4 Hidden Markov Models
• • 6.2.5 Principal Component Analysis
• • 6.2.6 Optimization algorithms
• • 6.3 Roles of statistical models in digital twins of materials processing
• • 6.4 Synergy between mechanistic and statistical models

Chapter 7: Sensing and control

• • 7.1 Introduction
• • 7.2 Sensors
• • 7.2.1 Temperature sensing
• • 7.2.2 Pressure measurements
• • 7.2.3 Flow sensors
• • 7.2.4 Vibration sensors
• • 7.2.5 Sensor data for control models
• • 7.3 Operations Research-based control models
• • 7.3.1 Linear programming
• • 7.3.2 Integer programming
• • 7.3.3 Dynamic programming
• • 7.3.4 Queuing theory
• • 7.3.5 Network flow models
• • 7.3.6 Markov decision processes
• • 7.3.7 Model predictive control
• • 7.3.8 Feedback and feed-forward control models
• • 7.4 Fuzzy Logic-based control models
• • 7.5 Data-driven control models
• • 7.6 Proportional-Integral-Derivative control models
• • 7.6.1 Advantages and disadvantages
• • 7.6.2 Types of proportional-integral derivative controllers
• • 7.7 Processing and storage of data for process control in manufacturing

Chapter 8: Digital twin implementation and case studies

• • 8.1 Introduction
• • 8.2 Implementation of digital twin
• • 8.2.1 Hardware and software integration
• • 8.2.2 Internet of things for connectivity
• • 8.2.3 Cyber physical systems in digital twins
• • 8.2.4 Validation and testing of the building blocks
• • 8.2.5 Uncertainty quantification for digital twins
• • 8.3 Examples of important applications
• • 8.3.1 A digital twin of additive manufacturing for part qualification
• • 8.3.2 A digital twin of fusion welding for weld quality control
• • 8.3.3 A digital twin of continuous die casting for quality control
• • 8.3.4 A digital twin for production control and planning
• • 8.3.5 A digital twin for controlling the microstructure of metallic parts

Chapter 9: Current status, research needs, and outlook

• • 9.1 Introduction
• • 9.2 Current status
• • 9.3 Research needs
• • 9.3.1 Data storage
• • 9.3.2 Blockchain
• • 9.3.3 Accessibility
• • 9.3.4 Cybersecurity
• • 9.3.5 Need for quantum computing
• • 9.3.6 High technology readiness level of building blocks
• • 9.3.7 Standardization
• • 9.4 Outlook
• • 9.4.1 Emerging trends
• • 9.4.2 Challenges and barriers to adoption
• • 9.4.3 Path forward