Search

The purpose of this paper is to present a new systematic procedure for synthesis of nonlinear controllers for single- input multiple-output systems. The procedure is based on several describing function models of the plant, and the nonlinear gains are obtained via an inverse describing function technique. The procedure and associated software are applied to control the bank angle of a cruise missile. The results are compared with a linear controller and another nonlinear controller that was previously reported in the open literature; it is shown that the developed procedure has resulted in a less conservative design (i.e., performance is not sacrificed to assure absolute stability). DOI: 10.1061/(ASCE)AS.1943-5525.0000012 CE Database subject headings: Control systems; Missiles; Nonlinear systems. Author keywords: Nonlinear control; Missile control; Bank angle control.

In this paper, a new systematic controller synthesis methodology for use with highly nonlinear multivariable and nonautonomous systems with application to a class of multivariable nonlinear aerospace systems is presented. The procedure is applied to a typical liquid propellant engine, and the performance of the resulting new control system is presented. In this research, the nonlinear dynamic model of the engine, which includes both soft and hard nonlinearities, is developed. The systematic controller design procedure is based on describing function models of the engine coupled with a new multivariable exact model matching procedure. Keywords: Describing Functions, Optimization, Design Methodologies, Fourier Integrals, Liquid Propellant Engine, Modeling, Simulation, Propulsion

Byline: Amir Nassirharand, Mohammad Hosain Alizadeh This paper aims at providing a solution to the problem of an autopilot design for non-agile aerospace vehicles with large length to diameter ratios. It is shown that quantitative feedback theory (QFT) may effectively be used to design an autopilot accounting for structural bending vibrations; this result is experimentally verified. Furthermore, it is demonstrated that a previously proposed factorisation-based autopilot design procedure for use with rigid aerospace vehicles may easily be adopted to arrive at lucid autopilots for flexible launch vehicles; this result is verified by a previously verified six degree-of-freedom simulation code. Although, both factorisation and QFT design approaches result in satisfactory autopilot systems, the factorisation design is simpler than that of QFT. The factorisation-based design man hour is considerably less than that of the QFT approach.

Byline: Amir Nassirharand, S.A. Hosseini This paper presents a new systematic procedure for decoupling and command tracking of multivariable, non-linear, and unstable systems. The design methodology is based on stabilisation of the multivariable system followed by generating its describing function models; two algebraic procedures for decoupling and tracking are used. Finally, the design must be verified by a non-linear simulation to make sure that approximations made during design are valid. The major outcome of this research is that controllers of a general class of non-linear systems may be designed taking into account the amplitude dependency features of that system. This amplitude dependency is the most important characteristic of a non-linear system. This is the first paper in the area of simultaneous decoupling and signal tracking of multivariable non-linear systems that is based on the application of the describing function technique and two other algebraic techniques. In fact, this is the first work in the area of decoupling of general unstable multivariable non-linear systems that is based on a describing function technique.

Byline: Amir Nassirharand The boundary layer conservation of mass, momentum, and energy equations are numerically solved. The inviscid pressure distribution is also numerically determined in order to specify the boundary conditions along the outer edge of the boundary layer. Finally, the Preston-tube pressure measurements are correlated to the corresponding theoretical skin friction coefficient values by means of a least-squares technique.

Byline: Amir Nassirharand A new controller synthesis algorithm for use with linear unstable plants is developed. The procedure is based on a factorisation theory. It involves determination of a desired transfer function followed by a two layer optimisation approach. The outer level non-linear optimisation layer optimises the assumed locations of the poles of the coprime factors of the plant and the coprime factors that are solutions to the Bezout identity. For a given pole locations of coprime factors, the inner layer optimisation (which is based on a previously developed factorisation procedure) is utilised to obtain a model-matching controller. The procedure is demonstrated by solving a difficult to control unstable plant, and results are compared with two other techniques previously reported in the open literature.