The fluid dynamics part of this project will cover two main components: computational fluid dynamics (CFD) and hydrodynamic testing.

Computational Fluid Dynamics: This component will integrate the hydrodynamic and motion control components. The hydrodynamic component will simulate fish movement while motion control will realise fish motion from one location to another. Pollutant spread in the water will also be modelled. The task involves following computational works:

  • Modelling pollutant spread due to current effect.
  • Modelling pollutant spread due to diffusion
  • Modelling pollutant spread due to sea wave
  • Modelling pollutant spread due to passing ships

Hydrodynamic Testing: The hydrodynamic testing programme will provide benchmarking data for the CFD study, and will also provide an understanding of the relationship between the manoeuvring behaviour of the fish with the swimming strategy employed (i.e. the amplitudes, frequencies and phases of the motions of the various sections). This will form the basis of a simplified engineering model of the motion control which will be used as a screening tool to identify key cases for more detailed analysis. Through tank (port, harbour) modelling of pollutant spread, results will give clear insight into how pollutant is spreading in the different environment or conditions. Based on this information, the SHOAL system can be designed with more accuracy and efficiency.

CFD is a high fidelity flow analysis and hydrodynamic performance estimation tool. Its earlier application was mainly on flow visualization, resistance prediction and performance optimisation in straight-ahead motion. Recently, its capability has been further extended to simulate ship manoeuvring motion and seakeeping behaviour. In the Virtual Manoeuvring Basin of the EU FP6 IP VIRTUE (a front-end research project on CFD), various approaches were developed to improve the reliability, accuracy and efficiency of numerical experimentation. These include hydrodynamic derivative approach under captive condition and direct numerical simulation solving RANS equations coupled to ship motion under free running condition.

Relatively few systematic tests of the hydrodynamics of robotic fish have been reported. Recent studies in the Acre Rd towing tank have included an extensive study of a robotic eel. In this study a simplified engineering model for the resistance, propulsion and side-forces was developed; the model was subsequently calibrated using tank test data in order to predict swimming speed, and to investigate several aspects of propulsion of robotic eels. The proposal CFD application is a hydrodynamic coefficient-based method supplemented by laboratory model tests. This is considered to be a more practical approach, with proven cost effectiveness, technical reliability and user-friendliness.

Some key factors to be considered in the computational modelling of these configurations are as follows.

  • Moving boundaries: hydrodynamic flows of fish are often associated with moving boundaries; these may be flapping fins or undulating bodies.
  • Two-way fluid-structure coupling: in many cases, the control surfaces (fins, appendages, etc.) are highly flexible and can undergo large deformations as a result of the hydrodynamic loading. This deformation can in turn have a significant effect on the flow, which can then modify the loading itself.
  • Unsteady flow mechanisms: the presence of moving and flexible control surfaces and/or the unsteady flow environment leads to configurations in which the dominance of unsteady flow mechanisms (added mass effects, dynamic stall, vortex shedding, vortex pairing, and vortex body and vortex-fin interactions) is a rule rather than an exception.

The proposed model test programme will address the novel challenges of adapting the well-understood techniques for establishing manoeuvring performance for rigid bodies (e.g. ships) to flexible swimming bodies. A customised planar motion (PMM) mechanism will be constructed for this purpose, based on previous experience with PMMs and other multiple degree of freedom motion generators. Key issues to be examined will include the relationship between manoeuvring performance and the amplitudes, frequencies and phases of the motion of the various segments of the robot.

The task to model how the movements of boats affect the shoal of robot fish is a complicated one. The wake of moving boats influences robot fish dynamically. This tricky problem will be studied through physical tank test and numerical simulation. The model tests in the towing tank will provide video image of the process. And computational simulation will give more information on the interaction of boats and robot fish.

The CFD modelling in this project will be the first successful application of cutting-edge CFD tools in this area. The technical challenge of the project derives from combined complexity of a deforming object, transient simulation, free sailing, coupling of control, and integration of other software, all of which require a most advanced computational technology.

The hydrodynamic testing programme will require the development of innovative techniques and strategies for measurement and analysis of manoeuvring performance of non-rigid bodies, and the relationship of the manoeuvring performance to the motion control strategies.