Skip to main content
main-content
Top

About this book

This book presents neural partial differentiation as an estimation algorithm for extracting aerodynamic derivatives from flight data. It discusses neural modeling of the aircraft system. The neural partial differentiation approach discussed in the book helps estimate parameters with their statistical information from the noisy data. Moreover, this method avoids the need for prior information about the aircraft model parameters. The objective of the book is to extend the use of the neural partial differentiation method to the multi-input multi-output aircraft system for the online estimation of aircraft parameters from an established neural model. This approach will be relevant for the design of an adaptive flight control system. The book also discusses the estimation of aerodynamic derivatives of rigid and flexible aircraft which are treated separately. The longitudinal and lateral-directional derivatives of aircraft are estimated from flight data. Besides the aerodynamic derivatives, mode shape parameters of flexible aircraft are also identified in the book as part of identification for the state space aircraft model. Since the detailed description of the approach is illustrated through the block diagram and their results are presented in tabular form with figures of parameters converge to their estimates, the contents of this book are intended for readers who want to pursue a postgraduate and doctoral degree in science and engineering. This book is useful for practicing scientists, engineers, and teachers in the field of aerospace engineering.

Table of Contents

Frontmatter

Chapter 1. Aircraft System Identification

Abstract
In a fairly complex system like aircraft, modeling and parameter estimation plays a crucial role in determining its stability and control characteristics.
Majeed Mohamed, Vikalp Dongare

Chapter 2. Neural Modeling and Parameter Estimation

Abstract
The neural modeling of a dynamic system is presented in this chapter. The former literature reported that ordinary differential equations can be solved by an neural-network-based approach (Lagaris et al. 1998).
Majeed Mohamed, Vikalp Dongare

Chapter 3. Identification of Aircraft Longitudinal Derivatives

Abstract
This chapter focuses on the application of neural networks to the problem of aircraft aerodynamic modeling and parameter estimation. The neural modeling and neural partial differentiation (NPD) method, which are presented in Chap. 2, are directly applied here to estimate the longitudinal dynamics of an aircraft system. Since the model of an aircraft system is established through the neural network, extends the use of NPD to multi-input and multi-output (MIMO) system ensures online estimation of each aircraft aerodynamics derivatives.
Majeed Mohamed, Vikalp Dongare

Chapter 4. Identification of Aircraft Lateral-Directional Derivatives

Abstract
The application of neural networks combined with partial differentiation of the neural outputs is discussed in this chapter to estimate lateral-directional flight stability and control derivatives from flight data. Primary investigation is carried out with simulated data and results are found to be encouraging to apply with flight data.
Majeed Mohamed, Vikalp Dongare

Chapter 5. Identification of a Flexible Aircraft Derivatives

Abstract
This chapter discusses the neural modeling of the flexible aircraft and how to extract the aerodynamic derivatives and structural mode shape parameters of flexible modes using the neural partial differentiation (NPD) method. The effects of flexibility on the flight dynamics of an aircraft have been shown to be quite significant, especially as the frequencies of its elastic modes become lower and approach those of the rigid body modes (Zerweckh et al. 1990; Meirovitch and Tuzcu 2001). The chapter discusses the neural modeling of the flexible aircraft and how to extract the aerodynamic derivatives and structural mode shape parameters of flexible modes using the NPD method. The effects of flexibility on the flight dynamics of an aircraft have been shown to be quite significant, especially as the frequencies of its elastic modes become lower and approach those of the rigid body modes (Colin et al. 2008; Majeed 2014). The characteristics of such flexible aircraft are altered significantly from those of a rigid aircraft, and the design of the flight control system may become drastically more complex (Bucharles and Vacher 2002). Therefore, mathematical modeling of a flexible aircraft for dynamic analysis and control system design is a major issue in flexible aircraft dynamics. The characteristics of such flexible aircraft are altered significantly from those of a rigid aircraft, and the design of the flight control system may become drastically more complex. Therefore, mathematical modeling of a flexible aircraft for dynamic analysis and control system design is a major issue in flexible aircraft dynamics.
Majeed Mohamed, Vikalp Dongare

Chapter 6. Conclusions and Future Work

Abstract
Neural partial differentiation (NPD) method is used for the parameter estimation. For this purpose initially, a neural model of multi-input and multi-output (MIMO) aircraft system is established. In addition to that, separate consideration of the longitudinal and lateral-directional dynamics simplifies the model considerably in terms of structure and the number of parameters.
Majeed Mohamed, Vikalp Dongare

Backmatter

Additional information

Premium Partner

    Image Credits