A Wind Simulation System for the Ocean Ranger Hydrodynamic Model Study

ABSTRACT During a severe storm on the night of February 14 to 15, 1982, the semi-submersible drilling platform, the Ocean Ranger, capsized and was lost. Shortly thereafter a Royal Commission was established to investigate the tragedy. AS part of the investigation the National Research Council of Canada (NRCC) was commissioned to carry out extensive model studies. The model studies involved measurement of the response of a hydrodynamic model of the Ocean Ranger under conditions which were designed to simulate the environmental conditions as they existed on the night of the capsize. For the simulation to be complete, both wave and wind loads had to be modeled. In order to simulate the wind loads on the hydrodynamic model, a sophisticated dynamic wind simulation system was developed. This paper deals with the design and analysis of that simulation system. A discussion of some of the results obtained using the simulation system is also included. INTRODUCTION The National Aeronautical Establishment of the NRCC was requested to conduct tests to determine the wind loads on the Ocean Ranger [Ref. 1]. Their tests involved measuring the forces and moments induced by the wind, for various drafts and orientations of a 1:100 scale aerodynamic model of the Ocean Ranger. These data were then used during the hydrodynamic tests which were carried out by the Hydraulic Laboratory of the NW using a 1:40 scale model. Wind loads may be applied to a hydrodynamic model using one of at least two different methods. The first method involves using a bank of fans to create a turbulent air flow over the hydrodynamic model. If the model is properly scaled aerodynamically, then the speed of the fans can be adjusted and modulated to achieve the desired wind loads. Alternatively, the distribution of tensions in lightweight lines attached to the model can be con-trolled to simulate the wind loads. The method using force controlled lines was selected for use by the Hydraulics Laboratory. A significant problem to be overcome was how to use the thousands of wind load measurements from the aerodynamic tests to predict the continuously changing wind loads on the hydrodynamic model. The method used to predict the wind loads was to develop a function for each component force and moment of the wind loads which required simulation during the hydrodynamic testing. Using only the aerodynamic test data, these functions were continually refined until, when applied to the aerodynamic test data, they were capable of producing a set of predicted wind loads which suitably matched the actual wind loads measured. The assumption was that if the functions were able to reproduce the wind loads measured using only the test setup specifications as input, then the functions would be capable of predicting wind loads on the hydrodynamic model provided that the input to the functions could be calculated. The analysis showed that the orientation of the vessel and the position of the vessel on a wave were required to calculate the inputs to the functions. This information was made available, in real-time, during the hydrodynamic testing.