Application of neural control to active structural load alleviation

Modern large transport aircrafts are subject to considerable structural flexibility during flight. In order to improve the dynamic response of such aircraft the use of active control technology is proposed which will alleviate the structural loads to which the airframe is subjected as a result of manoeuvre demands or encountering atmospheric turbulence. A number of methods, including linear optimal control theory, have been proposed as a solution of the structural load alleviation problem. This paper presents a method based on the use of intelligent neural controllers. Some comparisons of the effectiveness of different types of neural networks in this application are provided, and the resulting controlled responses obtained from digital simulation are compared with a baseline design based on linear quadratic regulator theory. The paper concludes with a recommendation for the use of one particular type of neural controller which provides superior performance for this application.<
>