Dynamic Modeling and Neural Network-Based Intelligent Control of Flexible Systems

Hejia Gao, PhD, is an Associate Professor at the School of Artificial Intelligence, Anhui University, Hefei, China. Previously, she was a Visiting Researcher at the Department of Mechanical, Industrial and Aerospace Engineering, Concordia University, Canada. She has published over 30 international journal and conference papers. Her research interests include neural networks, reinforcement learning, flexible systems, and vibration control. Wei He, PhD, is a Full Professor at the School of Automation and Electrical Engineering, University of Science and Technology Beijing, China. He has co-authored three books and published over 100 international journal and conference papers. He was awarded a Newton Advanced Fellowship from the Royal Society, UK, in 2017. His research interests include adaptive control, vibration control, and bionic flapping wing aircraft. Changyin Sun, PhD, is a Professor at the School of Automation, Southeast University, Nanjing, China. He has co-authored four books and published over 160 international journal papers. Prof. Sun is a Chinese Association of Automation Fellow. His research interests include intelligent control, flight control, pattern recognition, and optimal theory.

• About the Authors xiPreface xiiiAcknowledgments xviiAcronyms xix1 Introduction 11.1 Background and Motivation 11.2 Modeling and Control Strategies of Flexible Robotic Manipulators 51.3 Vibration Control Technologies of Flexible Building-like Structures 71.4 Modeling and Control Approaches of Bionic Flexible Flapping-wing Aircraft 81.5 Outline of the Book 92 Mathematical Preliminaries 132.1 Mathematical Preliminaries 132.1.1 Hamilton Principle 132.1.2 Model Discretization 142.1.2.1 Assumed Mode Method 142.1.2.2 Finite Rigid Element Method 142.1.3 Lagrange’s Equation Method 152.1.4 Neural Networks 152.1.5 Lyapunov Stability Theorem 162.1.6 Summary 183 Fuzzy Neural Network Control of the Single-Link Flexible Robotic Manipulator 193.1 Introduction 193.2 Problem Formulation 213.2.1 Dynamic Modeling 213.2.2 Model Discretization 223.3 Fuzzy Neural Network Control 243.3.1 Control Design 243.3.2 Stability Analysis 273.4 Numerical Simulations 303.4.1 Without Control 313.4.2 PD Control 323.4.3 Full-State Feedback 343.4.4 Output Feedback 343.5 Experimental Investigation 343.5.1 Experimental Testbed 343.5.2 Experimental Results 383.6 Summary 414 High-Gain Observer-Based Neural Network Control of the Two-Link Flexible Robotic Manipulator 434.1 Introduction 434.2 Problem Formulation 444.2.1 Dynamic Modeling 444.2.2 Model Discretization 474.3 High-Gain Observer-Based Neural Network Control 474.3.1 Control Design 474.3.2 Stability Analysis 494.4 Numerical Simulations 534.4.1 Simulation Results for Open-Loop System 534.4.2 Simulation Results for PD Control 544.4.3 Simulation Results for Neural Network Control 544.4.4 Comparison Between PD and NN Simulation Results 554.5 Experimental Investigation 584.5.1 Introduction of the Experimental Testbed 584.5.2 Experimental Results 584.5.3 Comparison Between PD and NN Experiment Results 614.6 Summary 625 Robust Adaptive Vibration Control for a String with Time-Varying Output Constraint 655.1 Introduction 655.2 Problem Formulation 675.2.1 Dynamics of the String System 675.2.2 Preliminaries 695.3 Control Design 695.3.1 Exact Model-Based Boundary Control 695.3.2 Robust Adaptive Boundary Control for System Parametric Uncertainty 725.4 The Solvability of the Inequality Equations 765.5 Numerical Simulations 815.6 Summary 846 Neural Network Vibration Control of a Stand-Alone Tall Building-Like Structure with an Eccentric Load 856.1 Introduction 856.2 Dynamic Modeling 886.2.1 Dynamic Modeling 886.2.2 Model Discretization 896.3 Neural Network Vibration Control 926.3.1 Control Design 926.3.2 Stability Analysis 936.4 Numerical Simulations 966.4.1 Simulation Parameters 966.4.2 Simulation Results 966.5 Experimental Investigation 1006.5.1 Introduction of the Experimental Testbed 1006.5.2 Experimental Results 1016.6 Summary 1057 Adaptive Vibration Control of a Flexible Structure Based on Hybrid Learning Controlled Active Mass Damping 1077.1 Introduction 1077.2 Dynamic Modeling 1097.3 Hybrid Learning Control 1137.3.1 Disturbance Observer Design 1137.3.2 Hybrid Learning Control Design 1157.3.3 Full-order State Observer 1187.4 Simulation Verification and Comparative Analysis 1187.5 Experimental Investigation 1207.5.1 Experimental Results of Passive Mode 1227.5.2 Experimental Results of PV Position Controller 1247.5.3 Experimental Results of HL Controller 1257.5.4 Comparisons and Discussions 1287.6 Summary 1298 Reinforcement Learning Control of a Single-Floor Building-Like Structure with Active Mass Damper 1318.1 Introduction 1318.2 Problem Formulation 1328.2.1 Dynamic Modeling 1328.2.2 Model Discretization 1348.3 Reinforcement Learning Control 1348.3.1 Control Design 1348.3.2 Stability Analysis 1368.4 Experimental Investigation 1378.5 Summary 1419 Disturbance Observer-Based Neural Network Control of a Flexible Flapping-Wing System 1439.1 Introduction 1439.2 Problem Formulation 1449.2.1 Dynamic Modeling 1449.2.2 Model Discretization 1469.3 Disturbance Observer-Based Neural Network Control 1489.3.1 Control Design 1489.3.2 Stability Analysis 1529.3.3 Simulation Results Without Control 1559.3.4 Simulation Results for PD Control 1559.3.5 Simulation Results for Full-State Feedback 1559.3.6 Simulation Results for Output Feedback 1589.4 Summary 15910 Adaptive Finite-Time Control of a Bionic Flexible Flapping-Wing Aircraft with Actuator Failures 16110.1 Introduction 16110.2 Problem Formulation 16310.2.1 Dynamic Modeling 16410.2.2 Model Discretization 16510.3 Adaptive Finite-Time Control 16710.3.1 Control Design 16710.3.2 Stability Analysis 16910.4 Numerical Simulations 17210.5 Summary 18111 Adaptive Vibration Control for Two-Stage Bionic Flapping Wings Based on Neural Network Algorithm 18311.1 Introduction 18311.2 Problem Formulation 18411.2.1 Dynamic Modeling 18411.2.2 Model Discretization 18511.3 Adaptive Vibration Control 18611.3.1 Control Design 18611.3.2 Stability Analysis 18811.4 Numerical Simulations 19111.5 Summary 19512 Boundary Vibration Control of a Floating Wind Turbine System with Mooring Lines 19712.1 Introduction 19712.2 System Modeling and Preliminaries 19912.2.1 Dynamical Model of Floating Wind Turbine Vibrations 20012.2.2 Preliminaries 20112.3 Controller Design 20212.4 Numerical Simulations 20612.5 Summary 21513 Conclusions 217References 219Index 243