An Experimental Investigation of Ride Control Algorithms for High-Speed Catamarans Part 1: Reduction of Ship Motions

Abstract

Ride control systems are essential for comfort and operability of high-speed ships, but it remains an open question what is the optimum ride control method. To investigate the motions of a 112 m high-speed catamaran fitted with a ride control system a 2.5 m model was tested in a towing tank. The model active control system comprised two transom stern tabs and a central T-Foil beneath the bow. Six ideal motion control feedback algorithms were used to activate the model scale ride control system and surfaces in a closed loop control system: heave control, local motion control and pitch control, each in a linear and nonlinear version. The responses were compared with the responses with inactive control surfaces and with no control surfaces fitted. The model was tested in head seas at different wave heights and frequencies and the heave and pitch Response Amplitude Operators (RAOs), Response Phase Operators (RPOs) and acceleration response were measured. It was found that the passive ride control system reduced the peak heave and pitch motions only slightly. The heave and pitch motions were more strongly reduced by their respective control feedback. This was most evident with nonlinear pitch control which reduced the maximum pitch RAO by around 50% and the vertical acceleration near the bow by about 40% in 60 mm waves. These reductions were influenced favorably by phase shifts in the model scale system which effectively contributed both stiffness and damping in the control action. Nomenclature CLα Control surface lift coefficient derivative (dCL/dα) Fr Froude number based on hull waterline length g Gravitational acceleration (m/s 2) H Model heave at LCG (m, positive up) H* Dimensionless model heave at LCG (heave/wave amplitude) l Length of model waterline (m) LCG Longitudinal Centre of Gravity LST Stern tab lift force (N) LTF T-Foil lift force (N) P Model pitch about LCG (radian, positive bow down) P* Dimensionless model pitch at LCG (pitch/wave slope) SST Stern tab planform area (m 2) STF T-Foil planform area (m 2) ST0 Stern tab initial angle of attack (radian, positive producing upward lift) TF0 T-Foil initial angle of attack (radian, positive producing upward lift) U Model forward speed (m/s) xST Distance between centre of pressure of the stern tabs and LCG (m) xTF Distance between centre of pressure of the T-Foil and LCG (m) αST Stern tabs angular deflection (radian, positive producing upward lift) αSTd Control system demand stern tabs angular deflection (radian, positive producing upward lift) αTF T-Foil angular deflection (radian, positive producing upward lift) αTFd Control system demand T-Foil angular deflection (radian, positive producing upward lift) ζ Wave amplitude (m) λ Wave length (m) ρ Water density (kg/m 3) ω0 Wave frequency in fixed coordinate (radian/s) ωe Wave encounter frequency (radian/s) ωe* Dimensionless wave encounter frequency

An Experimental Investigation of Ride Control Algorithms for High-Speed Catamarans Part 2: Mitigation of Wave Impact Loads

“…The principle here is simply that this action extracts maximum energy from any dynamic motion. As explained previously[7], the vertical velocities of the model at the longitudinal location of each control surface are O − 6 37 > and O + 6 ?3 > for the T-Foil and stern tab respectively. Thus in the local control mode we require control surface demands ; 37< = N(O − 6 37 >) and ; ?3< = ℎ(O + 6 ?3 >), where…”

“…The control surfaces lift coefficient derivatives (C Lα ) were determined from previous studies on the T-Foil [23] and stern tabs [25]. As previously explained [7] the T-Foil reaches the limit of its range before the stern tabs and therefore the parameter e is determined by Equation 8 in terms of the maximum T-Foil deflection and the parameter k is determined by Equation 9 in terms of e. Thus, the stern tabs operate at less than their full range of action.…”

“…Model testing in the towing tank for the present work was conducted as has been described previously [7]. The strain gauge data obtained during all towing tank tests undertaken on the catamaran model were analysed to identify the maximum centre bow entry force, the maximum total force and the slam force acting on the centre bow, and the maximum slam induced bending moments acting on the demi-hull.…”

“…Response Amplitude Operator (RAO) was reduced by up to 50% in 2.69 m (full scale) waves and the vertical acceleration near the bow by about 40% at the same model test conditions. In light of the ride control influence in reducing relative motions between the bow and water surface [7], it is clear that there are significant gains that can be achieved through the implementation of a ride control system with regard to reduction of wave impact slamming. This previous study thus establishes a basis for the current investigation which is to evaluate the influence of the ride control system on the structural loads during model tests undertaken in regular waves.…”

“…The model has a displacement of 28.3 kg, equivalent to a full scale displacement of 2545 tonnes, and an overall length of 2.5 m equivalent to the full scale length of 112 m.Figure 8of the previous paper[7] shows the model attached to the moving carriage. Data for model motions was derived from Linear Variable Differential Transformers (LVDTs) mounted on the forward and aft tow posts.…”