Advances in Evolutionary and Deterministic Methods for Design, Optimization and Control in Engineering and Sciences

• Main Authors: Gaspar-Cunha, António, Periaux, Jacques, Giannakoglou, Kyriakos C, Gauger, Nicolas R, Quagliarella, Domenico, Greiner, David
• Format: eBook
• Language: English
• Published: Cham Springer International Publishing AG 2020; Springer International Publishing
• Edition: 1
• Series: Computational Methods in Applied Sciences
• Subjects: Computational intelligence-Congresses; Engineering mathematics-Congresses; Mathematical optimization-Congresses
• ISBN: 3030574210; 9783030574215
• DOI: 10.1007/978-3-319-89988-6

This book presents improved and extended versions of selected papers from EUROGEN 2019,  a conference with interest on developing or applying evolutionary and deterministic methods in optimization of design and emphasizing on industrial and societal applications.

Table of Contents:

• 20 Neuroevolutionary Multiobjective Optimization of Injection Stretch Blow Molding Process in the Blowing Phase
• 12 Robustness Measures for Multi-objective Robust Design -- 12.1 Introduction -- 12.2 Robust Design -- 12.2.1 Robustness Measures -- 12.2.2 Expectation-Based Approach -- 12.2.3 Two-Phase Approach -- 12.2.4 Direct Approach -- 12.2.5 Uncertainty Quantification -- 12.3 Multi-objective Optimization -- 12.3.1 Constraint-Based Approach -- 12.3.2 Global Optimization Method -- 12.4 Aerodynamic Shape Optimization -- 12.5 Results for the Two-Phase Approach -- 12.6 Results for the Direct Approach -- 12.7 Summary and Outlook -- References -- 13 Uncertainty Assessment of an Optimized ERCOFTAC Pump -- 13.1 Introduction -- 13.2 Geometric Parameterization -- 13.2.1 Camber-Line -- 13.2.2 Thickness Function -- 13.2.3 Range of the Input Variables -- 13.3 CFD -- 13.3.1 Performance of the Real and Parametrized ERCOFTAC Pump -- 13.4 Optimization Strategy -- 13.5 Uncertainty Quantification Assessment -- 13.6 Results -- 13.6.1 Deterministic Optimization -- 13.6.2 Uncertainty Quantification Assessment -- 13.7 Conclusions -- References -- 14 Gradient-Based Aerodynamic Robust Optimization Using the Adjoint Method and Gaussian Processes -- 14.1 Introduction -- 14.2 Problem Definition -- 14.2.1 Deterministic Optimization -- 14.2.2 Robust Optimization -- 14.2.3 Parametrization -- 14.2.4 Uncertainties -- 14.3 Methodology -- 14.3.1 Numerical Solver -- 14.3.2 Adjoint Method -- 14.3.3 Surrogate Based Uncertainty Quantification -- 14.3.4 Optimization Framework -- 14.4 Results -- 14.4.1 Deterministic Optimization -- 14.4.2 Uncertainty Quantification -- 14.4.3 Robust Optimization -- 14.5 Conclusions -- References -- 15 A Multi Layer Evidence Network Model for the Design Process of Space Systems Under Epistemic Uncertainty -- 15.1 Introduction -- 15.2 Evidence Framework for Epistemic Uncertainty -- 15.3 Evidence-Based Robust Optimisation -- 15.4 Space Systems Project Life Cycle
• 8 The Effects of Crowding Distance and Mutation in Multimodal and Multi-objective Optimization Problems -- 8.1 Introduction -- 8.2 Related Work -- 8.3 Proposed NSGA-II-WSCD-NBM -- 8.3.1 Weighted-Sum Crowding Distance Method -- 8.3.2 Neighborhood Polynomial Mutation -- 8.4 Experimental Setting -- 8.4.1 Test Problems -- 8.4.2 Performance Measures -- 8.5 Analysis of Results -- 8.5.1 Influence of the Weight Values in WSCD -- 8.5.2 Influence of the Population Size in WSCD -- 8.6 Conclusion -- References -- 9 Combining Manhattan and Crowding Distances in Decision Space for Multimodal Multi-objective Optimization Problems -- 9.1 Introduction -- 9.2 Related Work -- 9.2.1 Crowding Distance in the NSGA-II Algorithm -- 9.3 Proposed NSGA-II-MDCD Algorithm -- 9.4 Experiments -- 9.4.1 Test Problems -- 9.4.2 Parameter Settings -- 9.4.3 Performance Measures -- 9.5 Analysis of Results -- 9.6 Conclusions -- References -- 10 An Unsteady Aerodynamic/Aeroacoustic Optimization Framework Using Continuous Adjoint -- 10.1 Introduction -- 10.2 Governing Equations -- 10.2.1 Flow Equations -- 10.2.2 Noise Prediction Using the FW-H Analogy -- 10.3 Formulation of the Continuous Adjoint Method -- 10.3.1 Constraint Imposition Methods -- 10.4 Verification of the Hybrid CFD/FW-H Solver -- 10.4.1 Monopole in Uniform Flow -- 10.4.2 Pitching Airfoil in Inviscid Flow -- 10.5 Optimization Results -- 10.6 Conclusions -- References -- 11 Discrete Adjoint Approaches for CHT Applications in OpenFOAM -- 11.1 Introduction -- 11.2 CHT Foundations -- 11.3 Algorithmic Differentiation -- 11.4 Checkpointing Considerations -- 11.5 Verifying the Checkpointing Implementation -- 11.6 Additional Considerations for Shape Optimization -- 11.7 CHT Sensitivity Results -- 11.8 Primal Copy Constructor Checkpointing -- 11.9 Summary and Outlook -- References
• 15.4.1 Pre-formulation -- 15.4.2 Formulation -- 15.4.3 Implementation -- 15.5 Multi-layer Evidence Network Model (ML-ENM) -- 15.6 Problem Formulation -- 15.7 Method -- 15.8 Results -- 15.9 Conclusion -- References -- 16 Solving Multi-objective Optimal Design and Maintenance for Systems Based on Calendar Times Using NSGA-II -- 16.1 Introduction -- 16.2 Methodology -- 16.2.1 Availability and Functionability Profile -- 16.2.2 Building Functionability Profiles -- 16.2.3 Multi-objective Optimization -- 16.3 Application Case -- 16.4 Results -- 16.5 Conclusions -- References -- 17 Assessment of Exergy Analysis of CFD Simulations for the Evaluation of Aero-Thermo-Propulsive Performance of Aerial Vehicles -- 17.1 Introduction -- 17.2 Why Exergy? -- 17.3 Formulation -- 17.4 Accuracy Assessment -- 17.4.1 Implementation -- 17.4.2 Sensitivity Analysis -- 17.5 Conclusion -- References -- 18 Surrogate-Based Shape Optimization of Centrifugal Pumps for Automotive Engine Cooling Systems -- 18.1 Introduction -- 18.2 Centrifugal Pump -- 18.3 Geometry Parametrization -- 18.3.1 Impeller -- 18.3.2 Volute -- 18.4 CFD Simulation -- 18.5 Optimization Strategy -- 18.6 Results and Discussion -- 18.7 Conclusions -- References -- 19 Towards an Open-Source Framework for Aero-Structural Design and Optimization Within the SU2 Suite -- 19.1 Introduction -- 19.2 Background -- 19.2.1 Structural FEM Solver -- 19.2.2 CFD Solver -- 19.2.3 Splining Method -- 19.2.4 Fluid Mesh Deformation Solver -- 19.2.5 Coupling Method -- 19.3 Application -- 19.3.1 Structural Solver Validation -- 19.3.2 Primal FSI Solver -- 19.4 Adjoint Based Optimization: Sensitivities Evaluation -- 19.4.1 Structural Sensitivities -- 19.4.2 FSI Sensitivities -- 19.5 Conclusions and Future Works -- References
• Intro -- Preface -- Contents -- 1 A Bi-Level Optimization Approach to Define Dynamic Tariffs with Variable Prices and Periods in the Electricity Retail Market -- 1.1 Introduction -- 1.2 Bi-Level Optimization Model -- 1.3 Hybrid Approaches -- 1.3.1 Free BL I -- 1.3.2 Free BL T -- 1.4 Experimental Results and Discussion -- 1.4.1 Results -- 1.5 Conclusions -- References -- 2 An Evolutionary Algorithm for a Bilevel Biobjective Location-Routing-Allocation Problem -- 2.1 Introduction -- 2.2 BB-LRA problem Formulation -- 2.3 EBA: An Evolutionary Biobjective Algorithm for Solving the BB-LRA Problem -- 2.3.1 Chromosome Encoding and Fitness Evaluation -- 2.3.2 Repairing a Chromosome -- 2.3.3 Initial Population -- 2.3.4 Population Handling: Crossover, Mutation and Selection -- 2.4 Computational Experiment -- 2.5 Conclusions -- References -- 3 Incorporation of Region of Interest in a Decomposition-Based Multi-objective Evolutionary Algorithm -- 3.1 Introduction -- 3.2 Background -- 3.2.1 Many-Objective Optimization -- 3.2.2 Introducing Preferences in MOEA -- 3.3 Methodology -- 3.3.1 The Preference Cone -- 3.4 Results and Discussion -- 3.4.1 Experimental Setup -- 3.4.2 ROI Definition -- 3.4.3 Discussion -- 3.5 Conclusions -- References -- 4 Solving Multiobjective Engineering Design Problems Through a Scalarized Augmented Lagrangian Algorithm (SCAL) -- 4.1 Introduction -- 4.2 Augmented Weighted Tchebycheff Methods -- 4.3 Augmented Lagrangian Technique Using the Hooke and Jeeves Pattern Search Method -- 4.4 Engineering Design Problems -- 4.4.1 Four-Bar Plane Truss -- 4.4.2 Cantilever Beam -- 4.4.3 Disc Break -- 4.4.4 I-Beam -- 4.4.5 Pressure Vessel -- 4.4.6 Speed Reducer -- 4.4.7 Two-Bar Truss -- 4.4.8 Welded Beam -- 4.4.9 Spring -- 4.4.10 Gear Train -- 4.5 Results and Discussion -- 4.6 Conclusions -- References
• 5 Many-Objective Multidisciplinary Evolutionary Design for Hybrid-Wing-Body-Type Flyback Booster on an Entirely Automated System -- 5.1 Introduction -- 5.2 Problem Definition -- 5.2.1 Objective Functions: 6 -- 5.2.2 Design Variables: 40 -- 5.2.3 Constraints: 5 -- 5.3 Full Automated Multidisciplinary Optimization System -- 5.3.1 Pre-process -- 5.4 Method of Numerical Functions -- 5.4.1 Optimizer -- 5.4.2 Data Mining -- 5.4.3 Evaluation Methods of Objective Functions -- 5.5 The Modifications of Problem Definitions in This Paper -- 5.5.1 The Geometrical Problems in the Previous Research -- 5.5.2 Modification Manner of the Geometrical Subjects -- 5.5.3 SBX Modification -- 5.6 Results and Discussion -- 5.6.1 Cause of the Extreme High Temperature Area -- 5.6.2 The Temperature Restraint on Boosters' Surface -- 5.6.3 Negative Expansion of Objective-Functions Space -- 5.7 Conclusions -- References -- 6 A Neuroevolutionary Approach to Feature Selection Using Multiobjective Evolutionary Algorithms -- 6.1 Introduction -- 6.2 Methodology -- 6.2.1 Classifier -- 6.2.2 Performance Measure for Classification -- 6.2.3 Multiobjective Optimization -- 6.3 Experimental Design -- 6.4 Results and Discussion -- 6.5 Conclusions -- References -- 7 Multi-objective Optimization in the Build Orientation of a 3D CAD Model -- 7.1 Introduction -- 7.2 Multi-objective Approach -- 7.2.1 Optimization Problem -- 7.2.2 Quality Measures -- 7.2.3 Multi-objective Genetic Algorithm -- 7.3 Experiments -- 7.3.1 Model -- 7.3.2 Implementation Details -- 7.3.3 Results for the SA versus SE Problem -- 7.3.4 Results for the SA versus BT Problem -- 7.3.5 Results for the SE versus BT Problem -- 7.3.6 Results for the SA versus SE versus BT Problem -- 7.3.7 Discussion of the Results -- 7.4 Conclusions and Future Work -- References