Data-Driven Evolutionary Modeling in Materials Technology

Due to efficacy and optimization potential of genetic and evolutionary algorithms, they are used in learning and modeling especially with the advent of big data related problems. This book presents the algorithms and strategies specifically associated with pertinent issues in materials science domain. It discusses the procedures for evolutionary multi-objective optimization of objective functions created through these procedures and introduces available codes. Recent applications ranging from primary metal production to materials design are covered. It also describes hybrid modeling strategy, and other common modeling and simulation strategies like molecular dynamics, cellular automata etc. Features: Focuses on data-driven evolutionary modeling and optimization, including evolutionary deep learning. Include details on both algorithms and their applications in materials science and technology. Discusses hybrid data-driven modeling that couples evolutionary algorithms with generic computing strategies. Thoroughly discusses applications of pertinent strategies in metallurgy and materials. Provides overview of the major single and multi-objective evolutionary algorithms. This book aims at Researchers, Professionals, and Graduate students in Materials Science, Data-Driven Engineering, Metallurgical Engineering, Computational Materials Science, Structural Materials, and Functional Materials.

Chapter 1: Introduction Chapter 2: Data with random noise and its modeling2.1 What is data-driven modeling2.2 Noise in the data2.3 Mitigating random noise in traditional manner2.4 Overfitting and underfitting problems2.5 Intelligent optimum models out of data with random noise Chapter 3: Nature inspired non-calculus optimization3.1 Using natural and biological analogues for modeling and optimization3.2 Replacing a gradient based optimization by directional evolutionary search and learning3.3 Binary encoding and Simple Genetic Algorithms3.4 The genetic operators in evolutionary algorithms3.5 Hamming cliff and Gray encoding3.6 Real encoding3.7 Tree encoding3.8 Sequence encoding3.9: Schema theorem Chapter 4: Single-objective evolutionary algorithms4.1 Preamble4.2 Simple Genetic Algorithm (SGA)4.3 Differential Evolution (DE)4.4 Particle Swarm Optimization (PSO)4.5 Ant Colony Optimization (ACO)4.6 Genetic Programming (GP)4.7 Micro Genetic Algorithm (µ-GA)4.8 Island Model of Genetic Algorithm4.9. Messy Genetic Algorithms4.10 Evolution Strategies (ES)4.11Cellular Automata4.12 Simulated Annealing4.13 Constraint handling4.14 Evolutionary algorithms as equation solver4.15 Evolutionary optimization of multimodal functions Chapter 5: Multi-objective evolutionary optimization 5.1 The notion of Pareto optimality5.2 The Pareto frontier and its representation5.3 Visualization of Pareto fronts5. 4 Pareto optimality vs Nash Equilibrium5.5 Ranking of non-dominated solutions5.6 Some special features of evolutionary multi-objective optimization algorithms5.7 Predator prey Genetic Algorithm5.8 Artificial Immune Algorithm5.9 Multi-objective Particle swarm optimization5.10 Nash Genetic Algorithm5.11 Algorithms for handling a large number of objectives5.12 The notion of k-optimality5.13 Reference Vector Evolutionary Algorithm (RVEA)5.14 Other prominent algorithms Chapter 6: Evolutionary learning and optimization using Neural Net paradigm6.1 Learning through conventional Neural Net6.2 Evolutionary Neural Net: the different possibilities6.3 EvoNN Algorithm: the learning module6. 4 EvoNN Algorithm: the module for assessing single variable response6.5 EvoNN Algorithm: the optimization module6.6 Pruning Algorithm Chapter 7: Evolutionary learning and optimization using Genetic Programming paradigm 7.1 Learning through single objective Genetic Programming7.2 Learning through Bi-objective Genetic Programming7.3 BioGP Algorithm: the learning module7.4 BioGP Algorithm: the optimization module7.5 BioGP Algorithm: the module for assessing single variable response7.6 Some special features of BioGP emphasized Chapter 8: The challenge of big data and Evolutionary Deep Learning8.1 The challenge of learning from big data8.2 The concept of Deep Neural Net8.3 Development of the EvoDN2 algorithm Chapter 9: Software available in public domain and the commercial software 9.1 Software for evolutionary data-driven modeling and optimization9.2 The commercial software modeFRONTIER9.3 The commercial software KIMEME9.4 Matlab versions of EvoNN, BioGP and EvoDN29.5 Running EvoNN in Matlab9.6 Running BioGP in Matlab9.7 Running EvoDN2 in Matlab9.8 Many objective optimization using cRVEA in Matlab9.9 Predictions using EvoNN/EvoDN2/BioGP models in Matlab9.10 Graphics support for using EvoNN/EvoDN2/BioGP models in Matlab9.11 Python versions of EvoNN, BioGP and EvoDN2 Chapter 10: Applications in Iron and Steel making 10.1 Evolutionary computation in Blast Furnace ironmaking10.2 Evolutionary optimization of the iron ore agglomeration processes10.3 Evolutionary optimization of the charging and burden distribution in blast furnace10.4 Evolutionary optimization of the blast furnace hot metal quality10.5 Evolutionary optimization of the blast furnace productivity, emission and cost of operation10.6 Some further analyses of the Si content blast furnace hot metal10.7 Many objective optimization of blast furnace10.8 The need for using a number of evolutionary algorithms in tandem in blast furnace optimization10.9 Some other evolutionary algorithms based studies related to blast furnace iron making10.10 Data-driven evolutionary algorithms applied to the alternate processes of ferrous production metallurgy10.11 Data-driven evolutionary optimization applied to the simulation of integrated steel plants10.12 Data-driven evolutionary studies for refining of steel10.13 Data-driven evolutionary algorithms in electric furnace steel making10.14 Evolutionary algorithms in continuous casting10.15 Single objective evolutionary algorithms based studies of continuous casting10.16 Multi-objective evolutionary algorithms based studies of continuous casting Chapter 11: Applications in chemical and metallurgical unit processing11. 1 Evolutionary optimization of chemical processing plants11. 2 Studies on the William and Otto Chemical Plant11.3 The process model for the William and Otto Chemical Plant11.4 Some more studies related to chemical technology11.5 Evolutionary optimization of primary metal production11.6 Evolutionary optimization of mineral processing11.7 Evolutionary optimization of aluminum extraction11.8 Evolutionary analysis applied to the thermodynamics of Pb-S-O vapor phase11.9 Evolutionary applied to applied to the leaching of ocean nodules and low grade ores11.10 A study on the Supported Liquid Membrane based separation11.11 Miscellaneous evolutionary studies in the area of hydrometallurgy11.12 Evolutionary algorithms in zone refining11.13 Few concluding remarks Chapter 12: Applications in Materials Design12.1 Data-driven evolutionary alloy design12.2 Evolutionary design of superalloys12.3 Evolutionary design of Aluminum alloys12.4 Evolutionary design of steels12.5 Evolutionary design of functional materials12.6 Evolutionary design of functionally graded materials12.7 Evolutionary design of biomaterials12.8 Evolutionary design of phase change materials12.9 Evolutionary design of some emerging and less common materials Chapter 13: Applications in Atomistic Materials Design13.1 Data-driven evolutionary atomistic material design13.2 Density functional theory13.3 Tight binding approximation13.4 Molecular dynamics simulations13.5 Empirical many body potential energy functions13.6 Development of empirical many body potentials using a data-driven evolutionary approach13.7 Data-driven evolutionary optimization of Fe-Zn system13.8 Evolutionary design of ionic materials13.9 Taylor-made evolutionary design of materials Chapter 14: Applications in Manufacturing14.1 Evolutionary algorithms in manufacturing14.2 Evolutionary optimization of rolling process14.3 Evolutionary optimization of forging14.4 Evolutionary optimization of extrusion14.5 Evolutionary optimization in welding14.6 Evolutionary optimization in sheet metal forming14.7 Evolutionary optimization in advanced particulate processing14.8 Evolutionary optimization of the heat treatment process14.9 Evolutionary studies on microstructure generation14.10 Evolutionary studies on metal and non-metal cutting Chapter 15: Miscellaneous Applications 15.1 Evolutionary algorithms in some specific applications15.2 Data-driven evolutionary algorithms applied to anisotropic yielding15.3 Data-driven evolutionary algorithms applied to battery design15.4 Evolutionary algorithms applied to VLSI design15.5 Evolutionary design of paper machine headbox15.6 Evolutionary algorithms in nucleic acid sequence alignment15.7 Evolutionary analysis of the heat transfer process in a bloom reheating furnaceEpilogueReferences

Professor Nirupam Chakraborti was educated in India and USA, receiving his B.Met.E from Jadavpur University, India, followed by an MS from New Mexico Tech, USA and PhD, PhD degrees from University of Washington, Seattle, USA. He joined Indian Institute of Technology, Kanpur as a member of the faculty in 1984 and switched to Indian Institute of Technology, Kharagpur in 2000. Internationally known for his pioneering work on evolutionary computation in the area of Metallurgy and Materials, globally, Professor Chakraborti was rated among the top 2% highly cited researchers in the Materials area in 2000, as per Scopus records. A former Docent of Åbo Akademi, Finland, former Visiting Professors of Florida International University and POSTECH, Korea, he also taught and conducted research at several other academic institutions in Austria, Brazil, Finland, Germany, Italy and the US. An international symposium, under the KomPlasTech 2019, which is world’s longest running conference series in the area of computational materials technology, was organized in Poland in 2019 to honor him. In 2020, an issue of a prominent Taylor of Francis journal, Materials and Manufacturing Processes was dedicated to him as well. In 2021 Indian Institute of Technology, Kharagpur and Indian Institute of Metals, a professional body, also organized another international seminar in his honor. This book is a culmination of Professor Chakarborti’s decades of research and teaching efforts in this area.