Optimal Design and Control of Multibody Systems: Proceedings of the IUTAM Symposium (IUTAM Bookseries, 42)

✍ Scribed by Karin Nachbagauer (editor), Alexander Held (editor)

• Publisher: Springer
• Year: 2024
• Tongue: English
• Leaves: 172
• Category: Library

✦ Synopsis

This book presents the proceedings of the IUTAM Symposium on Optimal Design and Control of Multibody Systems 2022, covering research papers in the realm of optimal structural and control design for both rigid and flexible multibody systems. It delves into the application of the adjoint approach, enabling the undertaking of extensive topology optimizations to unearth body designs that excel under time- and design-dependent loads. Encompassing presentations on (adjoint) sensitivity analysis, structural optimization, optimal control, robust optimization, artificial intelligence, machine learning, and computational methods and software development, the IUTAM Symposium 2022 showcased the latest breakthroughs and innovative methodologies. This book presents 14 meticulously peer-reviewed proceedings papers from the event, evenly split between the Optimal Design and Optimal Control panels.

✦ Table of Contents

Preface
Acknowledgement
Contents
Optimal Design
Optimization Processes for Automated Design of Industrial Systems
1 Conventional System Design
2 Why Classical Optimization Cannot Solve Industrial Design Problems
3 Strategies for Meeting Industrial Design Challenges
3.1 Gradient Avoidance Through Population-Based Optimization
3.2 More Practicability Through Multi-objective Optimization
3.3 More Efficiency Through Metamodeling
3.4 Constraint Avoidance Through Optimization-oriented Parametrization
3.5 Robust Design
3.6 Integration of Expert Knowledge via Supervised Learning
4 Conclusions and Outlook
References
Optimal Lightweight Engineering Design via a Three-Block Solver Scheme for Mechanical Analysis
1 Design Optimization as Enabler of Virtuous Circle of Lightweight Engineering Design
2 Modeling Mechanical Systems with Differential–Algebraic Equations
3 Design Sensitivity Analysis as Motor of Structural Design Optimization
4 Morphing Wing Demonstrator
5 Conclusion
References
Influence of Weighted Gradients in Topology Optimization of Flexible Multibody Systems
1 Introduction
2 Floating Frame of Reference Formulation
3 Objective Function and Gradient Calculation
4 Application Example
5 Modified SIMP Method
6 Optimization Results
7 Summary and Conclusion
References
Direct Differentiation of the Floating Frame of Reference Formulation via Invariants for Gradient-Based Design Optimization
1 Introduction
2 Flexible Multibody Dynamics Governing Equations with Invariants
3 Invariant-Based Sensitivity Analysis
4 Numerical Results
4.1 Comparison of Numerical and Analytical Differentiation
4.2 Design Optimization
5 Conclusion
References
Comparison of Continuous and Discrete Adjoint Methods for Topology Optimization in Structural Dynamics
1 Introduction
2 Structural Dynamics
2.1 Time Integration
3 Topology Optimization
4 Sensitivity Analysis
4.1 Continuous Adjoint Method
4.2 Discrete Adjoint Method
4.3 Modified Discrete Adjoint Method
5 Numerical Example
5.1 Optimization with the Modified Adjoint Method
6 Summary
References
Derivation of Geometrically Parameterized Shell Elements in the Context of Shape Optimization
1 Introduction
2 Theoretical Background
2.1 Finite Element Method
2.2 Model Order Reduction
3 Geometrically Parameterized Shell Element Formulation
4 Numerical Example
4.1 Modeling
4.2 Optimization
4.3 Validation
5 Conclusion and Outlook
References
Efficient Robust Topology Optimization of Eigenfrequencies Using the First-Order Second-Moment Method
1 Introduction
2 Robust Topology Optimization of Eigenfrequencies
2.1 Computation of Eigenfrequencies and Its Gradients
2.2 Robust Topology Optimization Framework
3 Computation of the Gradient of the Variance Using the First-Order Second-Moment Method
3.1 Direct Differentiation
3.2 Adjoint Method
3.3 Finite Differences
3.4 Principal Sensitivity FOSM
3.5 Comparison of Methods
4 Cantilever Beam Example
5 Conclusions
References
Optimal Control
Dynamic Analysis on Formation for the Flight Energy-Saving of a Flapping Wing Robots Flock
1 Introduction
2 Proposed Method
3 Simulations
3.1 A Single Flapping Wing Robot
3.2 Double Flapping Wing Robots
4 Discussion
5 Conclusions
References
Optimal Control of Open–Loop Multibody Systems Recovered from Data
1 Introduction
2 Multibody Dynamics in Joint Coordinates
3 Identification Based on Regularized Regression
3.1 Collect Data (Simulation, Measurements)
3.2 General Dynamics Model of MBS
3.3 Computing xi Coefficient with Regularized Regression
4 The Adjoint Method in Optimal Control
5 Optimal Control of Planar Open-Chain MBS
5.1 Identification
5.2 Optimal Control
6 Summary and Conclusions
References
Minimum Energy Control of Robot Manipulators Using a Simple Variable Stiffness Mechanism
1 Introduction
2 Design and Control of Energy-Saving Manipulator
3 Variable Stiffness Mechanism
4 Experimental Verification
5 Conclusions
References
Control Design of Machine Tool Feed Drives Using Mechatronic System Simulation
1 Introduction
2 Mechatronic System Simulation
2.1 Finite Element Model
2.2 Control Loop Model
2.3 Exact Feedforward and Trajectory Planning
3 Application, Results and Discussion
4 Conclusion
References
Feedforward Control for a Manipulator with Flexure Joints Using a Lagrangian Neural Network
1 Introduction
2 Method
2.1 2-DOF Manipulator with Flexure Joints
2.2 White-Box Model
2.3 Lagrangian Neural Network (LNN)
3 Results
3.1 Training Data
3.2 White-Box Identification
3.3 LNN Training
3.4 FNN Training
3.5 Offline Force Estimates
3.6 Online Feedforward Control
4 Conclusions
References
On the Numerical Stability of Discretised Optimal Control Problems
1 Introduction
2 Optimal Control Problem: Indirect Method
3 Motivating Example: Linear OCP
3.1 Mid-Point Scheme
3.2 Implicit Euler Scheme
3.3 Symplectic Euler Scheme
4 Nonlinear OCP: Inverted Elastic Pendulum
5 Summary
References
On the Usage of Analytically Computed Adjoint Gradients in a Direct Optimization for Time-Optimal Control Problems
1 Introduction
2 Use of Adjoint Variables in Direct Optimization Approaches
3 Computation of First-Order Derivatives
3.1 Direct Differentiation Approach for Discrete Control Parameterization
3.2 Adjoint Gradient Approach for Discrete Control Parameterization
3.3 Discussion on Duality of Gradients
4 Numerical Example
4.1 Task Description and Optimization Problem
4.2 Results
5 Conclusions
References
Author Index