We study the fluid dynamics of two fish-like bodies with synchronised swimming patterns. Our studies are based on two-dimensional simulations of viscous incompressible flows. We distinguish between motion patterns that are externally imposed on the swimmers and self-propelled swimmers that learn manoeuvres to achieve certain goals. Simulations of two rigid bodies executing pre-specified motion indicate that flow-mediated interactions can lead to substantial drag reduction and may even generate thrust intermittently. In turn we examine two self-propelled swimmers arranged in a leader-follower configuration, with a-priori specified body-deformations. We find that the swimming of the leader remains largely unaffected, while the follower experiences either an increase or decrease in swimming speed, depending on the initial conditions. Finally, we consider a follower that synchronises its motion so as to minimise its lateral deviations from the leader's path. The leader employs a steady gait while the follower uses a reinforcement learning algorithm to adapt its swimming-kinematics. We find that swimming in a synchronised tandem can yield up to about 30% reduction in energy expenditure for the follower, in addition to a 20% increase in its swimming-efficiency. The present results indicate that synchronised swimming of two fish can be energetically beneficial.

We present MRAG-I2D, an open source software framework, for multiresolution simulations of two-dimensional, incompressible, viscous flows on multicore architectures. The spatiotemporal scales of the flow field are captured by remeshed vortex methods enhanced by high order average-interpolating wavelets and local time-stepping. The multiresolution solver of the Poisson equation relies on the development of a novel, tree-based multipole method. MRAG-I2D implements a number of HPC strategies to map efficiently the irregular computational workload of wavelet-adapted grids on multicore nodes. The capabilities of the present software are compared to the current state-of-the-art in terms of accuracy, compression rates and time-to-solution. Benchmarks include the inviscid evolution of an elliptical vortex, flow past an impulsively started cylinder at  and simulations of self-propelled anguilliform swimmers. The results indicate that the present software has the same or better accuracy than state-of-the-art solvers while it exhibits unprecedented performance in terms of time-to-solution.

Undulatory locomotion is an archetypal mode of propulsion for natural swimmers across scales. Undulatory swimmers convert transverse body oscillations into forward velocity by a complex interplay between their flexural movements, morphological features and the fluid environment. Natural evolution has produced a wide range of morphokinematic examples of undulatory swimmers that often serve as inspiration for engineering devices. It is, however, unknown to what extent natural swimmers are optimized for hydrodynamic performance. In this work, we reverse-engineer the morphology and gait for fast and efficient swimmers by coupling an evolution strategy to three-dimensional direct numerical simulations of flows at intermediate Reynolds numbers. The fastest swimmer is slender with a narrow tail fin and performs a sequence of C-starts to maximize its average velocity. The most efficient swimmer combines moderate transverse movements with a voluminous head, tapering into a streamlined profile via a pronounced inflection point. These optimal solutions outperform anguilliform swimming zebrafish in both efficiency and speed. We investigate the transition between morphokinematic solutions in the speed–energy space, laying the foundations for the design of high-performance artificial swimming devices.

Undulatory swimmers flex their bodies to displace water, and in turn, the flow feeds back into the dynamics of the swimmer. At moderate Reynolds number, the resulting flow structures are characterized by unsteady separation and alternating vortices in the wake. We use the flow field from simulations of a two-dimensional, incompressible viscous flow of an undulatory, self-propelled swimmer and detect the coherent Lagrangian vortices in the wake to dissect the driving momentum transfer mechanisms. The detected material vortex boundary encloses a Lagrangian control volume that serves to track back the vortex fluid and record its circulation and momentum history. We consider two swimming modes: the C-start escape and steady anguilliform swimming. The backward advection of the coherent Lagrangian vortices elucidates the geometry of the vorticity field and allows for monitoring the gain and decay of circulation and momentum transfer in the flow field. For steady swimming, momentum oscillations of the fish can largely be attributed to the momentum exchange with the vortex fluid. For the C-start, an additionally defined jet fluid region turns out to balance the high momentum change of the fish during the rapid start.

We study a self-propelling pair of steadily counter-rotating cylinders in simulations of a two-dimensional viscous fluid. We find two strikingly, opposite directions for the motion of the pair that is characterized by its width and rotational Reynolds number. At low Reynolds numbers and large widths, the cylinder pair moves similarly to an inviscid point vortex pair, while at higher Reynolds numbers and smaller widths, the pair moves in the opposite direction through a jet-like propulsion mechanism. Increasing further the Reynolds number, or decreasing the width, gives rise to non-polarised motion governed by the shedding direction and frequency of the boundary-layer vorticity. We discuss the fundamental physical mechanisms for these two types of motion and the transitions in the corresponding phase diagram. We discuss the fluid dynamics of each regime based on streamline plots, tracer particles, and the vorticity field. The counter rotating cylinder pair serves as a prototype for self-propelled bodies and suggests possible engineering devices composed of simple components and tunable by the rotation and width of the cylinder pair.

Vortical flows are crucial in applications ranging from aircraft takeoff to fish swimming. Numerical flow simulations can increase our understanding of such phenomena in order to improve the performance of engineering devices and characterize natural processes. Yet simulations of unsteady 3D flows on modern computer architectures face several challenges. The accuracy and validity of the numerical methods need to be established before physical processes can be predicted with confidence. Efficient simulations demand high-performance software that can adapt computational elements according to the spatiotemporal scales in the flow. Physical relevance is obtained only when the resulting flow data is translated into knowledge of the dominating structures and interactions through analysis and post-processing. In this work, we contribute to each of these issues. We use the remeshed vortex method for the simulation of selected 3D flow cases. We validate the method with vortex dynamics computations at various Reynolds numbers, and show that similar accuracy to the established pseudo-spectral method can be achieved. The solver is then applied to vortex reconnection, a mechanism for energy decay in turbulent flows, at an unprecedented Reynolds number of 10⁴. An analysis of the results uncovers novel flow features such as axial flow generation and secondary reconnection instabilities. The solver is extended with fluid-structure interaction capabilities and coupled to an optimization algorithm to create a versatile reverse-engineering framework. This approach is then used to identify the optimally fast and efficient morphologies and kinematics of 3D undulatory swimmers. The optimal solutions outperform a natural reference swimmer by reaching almost three times higher speeds and over four times higher eciencies. We investigate how the optimal swimmers manipulate the fluid to accomplish these feats. Lastly, we develop a 3D multiresolution remeshed vortex method to improve the computational eciency of fluid-structure interaction simulations. It is designed for high performance on modern compact compute platforms and employs a hybrid CPU/GPU velocity solver. The software enables the first ever computation multiple 3D self-propelled swimmers and their hydrodynamic interactions, achieved using only four compute nodes.

We present a multipole-based N-body solver for 3D multiresolution, block-structured grids. The solver is designed for a single heterogeneous CPU/GPU compute node, and evaluates the multipole expansions on the CPU while offloading the compute-heavy particle-particle interactions to the GPU. The regular structure of the destination points is exploited for data parallelism on the CPU, to reduce data transfer to the GPU and to minimize memory accesses during evaluation of the direct and indirect interactions. The algorithmic improvements together with HPC techniques lead to 81% and 96% of the upper bound performance for the CPU and GPU parts, respectively.

We investigate the optimal morphologies for fast and efficient anguilliform swimmers at intermediate Reynolds numbers, by combining an evolution strategy with three-dimensional viscous vortex methods. We show that anguilliform swimmer shapes enable the trapping and subsequent acceleration of regions of fluid transported along the entire body by the midline travelling wave. A sensitivity analysis of the optimal morphological traits identifies that the width thickness in the anterior of the body and the height of the caudal fin are critical factors for both speed and efficiency. The fastest swimmer without a caudal fin, however, still retains 80 % of its speed, showing that the entire body is used to generate thrust. The optimal shapes share several features with naturally occurring morphologies, but their overall appearances differ. This demonstrates that engineered swimmers can outperform biomimetic swimmers for the criteria considered here.

We investigate the C-start escape response of larval fish by combining flow simulations using remeshed vortex methods with an evolutionary optimization. We test the hypothesis of the optimality of C-start of larval fish by simulations of larval-shaped, two- and three-dimensional self-propelled swimmers. We optimize for the distance travelled by the swimmer during its initial bout, bounding the shape deformation based on the larval mid-line curvature values observed experimentally. The best motions identified within these bounds are in good agreement with in vivo experiments and show that C-starts do indeed maximize escape distances. Furthermore we found that motions with curvatures beyond the ones experimentally observed for larval fish may result in even larger escape distances. We analyse the flow field and find that the effectiveness of the C-start escape relies on the ability of pronounced C-bent body configurations to trap and accelerate large volumes of fluid, which in turn correlates with large accelerations of the swimmer.

We present simulations of the long-time dynamics of two anti-parallel vortex tubes with and without initial axial flow, at Reynolds numberRe = Γ/ν = 10⁴. Simulations were performed in a periodic domain with a remeshed vortex method using 785 × 10⁶ particles. We quantify the vortex dynamics of the primary vortex reconnection that leads to the formation of elliptical rings with axial flow and report for the first time a subsequent collision of these rings. In the absence of initial axial flow, a −5/3 slope of the energy spectrum is observed during the first reconnection of the tubes. The resulting elliptical vortex rings experience a coiling of their vortex lines imparting an axial flow inside their cores. These rings eventually collide, exhibiting a −7/3 slope of the energy spectrum. Studies of vortex reconnection with an initial axial flow exhibit also the −7/3 slope during the initial collision as well as in the subsequent collision of the ensuing elliptical vortex rings. We quantify the detailed vortex dynamics of these collisions and examine the role of axial flow in the breakup of vortex structures.

We present a vortex particle method coupled with a penalization technique to simulate single and multiple swimmers in an incompressible, viscous flow in two and three dimensions. The proposed algorithm can handle arbitrarily deforming bodies and their corresponding non-divergence free deformation velocity fields. The method is validated on a number of benchmark problems with stationary and moving boundaries. Results include flows of tumbling objects and single and multiple self-propelled swimmers.

We present a validation study for the hybrid particle-mesh vortex method against a pseudo-spectral method for the Taylor–Green vortex at Re_Γ = 1600 as well as in the collision of two antiparallel vortex tubes at Re_Γ = 10,000. In this study we present diagnostics such as energy spectra and enstrophy as computed by both methods as well as point-wise comparisons of the vorticity field. Using a fourth order accurate kernel for interpolation between the particles and the mesh, the results of the hybrid vortex method and of the pseudo-spectral method agree well in both flow cases. For the Taylor–Green vortex, the vorticity contours computed by both methods around the time of the energy dissipation peak overlap. The energy spectrum shows that only the smallest length scales in the flow are not captured by the vortex method.

In the second flow case, where we compute the collision of two anti-parallel vortex tubes at Reynolds number 10,000, the vortex method results and the pseudo-spectral method results are in very good agreement up to and including the first reconnection of the tubes. The maximum error in the effective viscosity is about 2.5% for the vortex method and about 1% for the pseudo-spectral method. At later times the flows computed with the different methods show the same qualitative features, but the quantitative agreement on vortical structures is lost.