Feedforward Control Law For A Shipboard Crane With Maryland Rigging System

Abstract

A state-space model of the Maryland Rigging shipboard crane is derived from Newton's law under the assumptions of boom stiffness, fully controllable boom motion, no cable elasticity, no damping, and full control authority for changing the length of the rope. A chaotic rolling moment, with a dominant frequency of the same order as the resonance frequency of the shipboard crane, is applied to the ship as an external disturbance. The effect of this disturbance is studied. Since designing a controller by means of analytical methods for this system is too complex, we use a novel approach to this problem that focuses on the equilibrium point. By deriving the equations for calculating the position of the equilibrium point of the load in space, we change the problem to minimizing the change in the position of this point. A feedforward type controller is then designed as to keep the load closest to the “equilibrium point” for the actual roll angle. The controller seeks to suppress the load sway caused by the ship's rolling motion by changing the luffing angle while the friction in the pulley is assumed to be negligible. Changing the luffing angle seems to be the most effective control action in shipboard cranes. The feedforward gain is then optimized by numerical methods. The simulation results for this controller show a huge decrease in the sway magnitude as compared to the cases with no control. The roll angle, luffing angle of the boom, and the length of the rope are changed individually and then the related optimum feedforward gains are numerically obtained. Using these data, the mapping of the optimum gain based on these variables is derived. Scheduling the gain based on this mapping greatly improves the performance of the feedforward controller. This procedure can be repeated for similar applications.

An application of model predictive control for a shipboard crane

“…We also chose the sampling frequency to be one second and the control horizon Tu = 50 seconds. The details of the feedforward component of the control system can be found in (Kimiaghalam et al 2000). It suffices to say that we have used a fixed gain of 1.67 in the feedforward component of our controller.…”

“…The detailed description is shown in (Kimiaghalam, et al 2000). Our basic control objective is to change b in order to minimize or eliminate the displacement of the load in x direction.…”

“…If finding a closed loop solution is not affordable, an open loop solution can be used as a fair alternative. Feedforward control (Kimiaghalam et al 2000) for example shows an outstanding result given that we have an accurate model of the plant to design the feedforward control law with a scheduled gain and accurate sensors to measure the input disturbance. Feedforward control is easy to design and implement, and it can provide great reduction in error caused by input disturbance(s).…”

Modeling and Control of a Bifilar Crane Payload

“…Based on [6], it can be concluded that the research on crane payloads modeled as a simple spherical pendulum is dominating in the field. However, in [7] the authors model a planar Maryland Rigging system and derive a control law for the sway motion. A Maryland Rigging system is a crane payload configuration 1 The authors are with the Dept.…”

Nonlinear coordination control of offshore boom cranes with bounded control inputs

“…[17][18][19][20][21][22] An intuitive idea to improve the performance of offshore cranes is to change their mechanical structures. For example, the well-known Maryland rigging system 23 is intensively employed to attenuate the payload swing. Although some improvements are achieved, these structure modifications can only conquer the disturbances passively, and their performance cannot be theoretically guaranteed.…”