"Synthesis of a Heave Control System for an Amphibious Hovercraft"

Abstract

The need for improved 'ride' quality of amphibious hovercraft, both in commercial and military applications, is widely accepted. Lack of passenger ride comfort is likely to inhibit the successful development of craft of longer range than the present, and inability to provide a sufficiently stable platform for weapon deployment will similarly limit the use of hovercraft in this application. Apart from increases in craft size, the principal weans to improve ride quality is through the control of the heaving motion and, in particular, reducing the magnitude of the heave acceleration experienced whislt traversing water or other surface roughness. Various solutions to this question of heave control have been proposed, but very few have been developed to a satisfactory state. In our work at C.I.T., Cranfield, emphasis has been on the development of an 'active-fan' system in which an axial-flow, lift-fan has its characteristics modified by means of fan blades whose pitch angle, to the plane of rotation, is continuously varied during operation. The angular movement of the blades is obtained by means of an hydraulic actuator whose action is controlled by means of feedback loops of craft cushion pressure (inner loop) and heave acceleration(outer loop). During the design and development of this system considerable difficulty was experienced in defining appropriate transfer function to describe the craft-cushion dynamics, these being nonlinear in character. To overcome this problem resort was made to parameter identification methods using nonlinear optimization algorithms. This approach proved to be very satifactory one and was, subsequently, extended to other sub-systems of the craft control system as well as to analysis of the overall response of the craft when tested over waves. These tests were conducted over solid waves on the wall of the Cranfield Whirling-Arm Facility. As a result of the above tests and related modelling, it has been possible to formulate an optimal control strategy,this in conjunction with a model of the Pierson-Moskowitz spectrum of sea state. The control optimization algorithm used is the modified Wiener filter theory in which is included a specific criterion to ensure passenger ride comfort. Simulated results, which have been obtained through the use of the Cranfield Advanced Continuous Simulation Language, are encouraging and ripe for further development.