No abstract

The formulation, tuning and performance of a signal prediction algorithm as applied to the determination of a Go-NoGo state are discussed. Simulations were used to tune and assess the performance of the signal prediction algorithm. The paper describes the development of useful criteria, based on the mean and standard deviation of the predicted signal, used for producing the Go-NoGo state. A latching algorithm was used to improve the output of the Go-NoGo state.

Improvements to launch and recovery operations at sea are driven by the desire to increase safety and operational availability. This thesis presents various tools to improve motion compensation strategies in maritime launch and recovery: a 3D computer simulator to examine wave synchronization, a signal prediction algorithm for Go-NoGo states and a hardware setup to simulate ship and wave motion. The 3D simulator of towed body dynamics was advanced to model the wave interactions with the cable and towed body as the body exits the water. Small scale simulations were run to investigate the inclusion of wave synchronization in established active heave compensation strategies where the hypothesis that wave synchronization would reduce variations in cable tension was not supported; the simulations demonstrated that wave synchronization increased variations in cable tension compared to simulations not using motion compensation. The use of a signal prediction method that forecasts a periodic signal based only on historic data of the signal was explored. The method is a means to predict safe breach events where the prediction algorithm was advanced and tuned to determine Go-NoGo states. A Go scenario identified by the mean and one standard deviation of the predicted signal was found to produce a Go-NoGo signal that agreed most with the desired Go-NoGo signal for forecasts up to 10 s. For the development of laboratory equipment for the Carleton University flume tank, a ship motion simulator was designed and built to emulate 5 degrees-of-freedom of ship motion and a kinematic analysis was performed to characterize the system workspace. For producing waves in the flume tank, a design methodology was developed for the design of a plunger-type wavemaker. A numerical model for determining the wave amplitude to actuator stroke length ratio was advanced to include the effects of a flow current. The design methodology enables the designer to select an appropriate actuator and plunger shape based on an operating point that incorporates multiple design variables.

No abstract

No abstract