Neural Networks Modelling and Control: Applications for Unknown Nonlinear Delayed Systems in Discrete-Time

Neural Networks Modelling and Control: Applications for Unknown Nonlinear Delayed Systems in Discrete-Time focuses on modelling and control of discrete-time unknown nonlinear delayed systems under uncertainties based on Artificial Neural Networks. First, a Recurrent High Order Neural Network (RHONN) is used to identify discrete-time unknown nonlinear delayed systems under uncertainties, then a RHONN is used to design neural observers for the same class of systems. Therefore, both neural models are used to synthesize controllers for trajectory tracking based on two methodologies: sliding mode control and Inverse Optimal Neural Control. The designed neural controllers are applied in real-time first to a linear induction motor prototype and then to a networked robotic system. As well as considering the different neural control models and complications that are associated with them, this book alsoanalyzes potential applications, prototypes and future trends. In-depth analysis of neural control models and methodologiesComprehensive review of common problems in real-life neural network systemsAnalysis of potential applications, prototypes and future trends INDICE: 1. Introduction1.1. Time-delay system1.2. System model1.3. Neural Identification1.4. Neural state observer1.5. Neural block control 1.5.1. Discrete-time Sliding mode control1.5.2. Inverse optimal control1.6. Problem Statement1.7. Objectives1.8. Background information1.9. Book Structure2. Mathematical preliminaries2.1. Time-delay systems2.1.1. Time-delay2.1.2. Time-delay system2.1.3. Nonlinear discrete-time system with time-delays2.2. Recurrent high order neural network 2.2.1. Discrete-time recurrent high order neural network2.2.2. Extended Kalman Filter based training for recurrent high order neural networks3. Recurrent high order neural network identification of nonlinear discrete-time unknown system with time-delays.3.1. System identification3.2. Neural Identification3.3. Design of a neural identifier based on a recurrent high order neural network for a nonlinear discrete-time unknown system withtime-delays.3.4. Simulation results of the recurrent high order neural network identifier3.4.1. Van der Pol oscillator3.4.2. Differential Robot4. Neural identifier-control scheme for nonlinear discrete-time unknown system with time-delays4.1. Neural identifier-control scheme, discrete-time sliding modes4.1.1. Discrete-time sliding mode control4.1.2. Real-time results of the neural identifier-control scheme using sliding mode control4.1.2.1. Linear Induction motor with time-delays, test 14.1.2.2. Linear Induction motor with time-delays, test 24.1.2.3. Linear Induction motor with time-delays, test 34.2. Neural identifier-control scheme, inverse optimal control4.2.1. inverse optimal control4.2.2. Real-time results of the neural identifier-control scheme using inverse optimal control4.2.2.1. Application to a differential robot4.2.2.1.1. Differential robot, test 14.2.2.1.2. Differential robot, test 25. Recurrent high order neural network observer of nonlinear discrete-time unknown systems with time-delays.5.1. Neural observer5.2. Design of a full order neural observer based on a recurrent high order neural network for a nonlinear discrete-time unknownsystem with time-delays.5.2.1. Simulation results of the recurrent high order neural network full order observer5.3. Design of a reduced order neural observer based on a recurrent high order neural network for a nonlinear discrete-timeunknown system with time-delays.5.3.1. Simulation results of the recurrent high order neural network reduced order observer6. Neural observer-control scheme for nonlinear discrete-time unknown system with time-delays6.1. Design of a reduced order neural observer based on a recurrent high order neural network for a nonlinear discrete-timeunknown system with time-delays. 6.1.1. Simulation results of the neural observer-control6.1.2. Real-time results of the neural observer-control7. Concluding remarks and future trendsAppendixA. Artificial neural networksa. Biological neural networksi. Biological neuronii. Biological synapseiii. Classification of neuronsb. Artificial neural networksc. Activation functionsd. Artificial neural networks classificationi. Single-layer neural networksii. Multilayer neural networksiii. Recurrent neural networksB. Linear induction motor prototypea. Linear induction motorb. How does a linear induction motor work?c. Linear induction motor modeld. Flux observere. Linear induction motor prototypei. Electric drive by induction motorii. Linear induction motor prototypeiii. Prototype del robot differentialC. Differential robot prototypea. All-terrain tracked robotb. All-terrain tracked prototype

• ISBN: 978-0-12-817078-6
• Editorial: Academic Press
• Encuadernacion: Rústica
• Páginas: 260
• Fecha Publicación: 01/01/2020
• Nº Volúmenes: 1
• Idioma: Inglés