Time Series Prediction for Generalized Heave Displacement of a Shipborne Helicopter Platform

Abstract

A shipborne helicopter platform, which is connected with a ship by hydraulic cylinders, is helpful for a helicopter to land and take off safely. The large inertia of the system arouses the problem of time delay when the system running. In order to solve that, the short time prediction for the generalized heave motion (the coupling result of roll, pitch and heave) of the platform is needed. In this paper, an automation regressive moving average (ARMA) model is proposed to do the prediction. A least square (LS) algorithm is commonly used to estimate the parameters of the model. However, the parameter estimation based on LS algorithm is easy to drift and be unstable when the system has random noise. To improve the problem, a damped recursive least square (DRLS) algorithm is introduced to estimate the parameters of the ARMA model. Using the collected real time data, the simulations suggest that the DRLS algorithm is able to increase the stability of parameter estimation and the ARMA model can get a multi-step prediction for generalized heave displacement of a shipborne helicopter platform.