In this work, the turning motion of fish is numerically studied. It aims at finding out the factors that could affect the turning performance of fish. A NACA0012 airfoil is selected as a simplified yet representative model. The deformation of fish body is prescribed and its turning motion is fully induced by the fluid-structure interaction (FSI). Numerical simulations are carried out under a series of operating conditions by varying the prescribed motion amplitude, frequency, and angular velocity of the fish. Results show that the turning motion of the fish head is affected by a few factors. A greater angular velocity leads to a smaller turning radius. Moreover, reducing the frequency of the fish tail and increasing the undulating amplitude will also reduce the turning radius.
At present, propellers are used for most of the underwater vehicles. It has some disadvantages such as large noise and low flexibility. The new fish-like underwater robots could learn from fish make use of complicated vortices and adapt their shape to achieve optimal outcomes instinctively. Therefore, it is interesting to learn how a fish could avoid unexpected obstacles with few efforts.
High maneuverability of fish shows up in its gyration radius, which is less than one body position. The hydrodynamic performance of fish turning was first examined by Weihs (1972). He modelled fish turning using the Lighthill's Large-Amplitude Slender-Body Theory. Sakakibara, Nakagawa and Yoshida (2004) used Stereo-PIV to measure the three-component velocity distribution around a live fish, and the side jet shed against the swept tail during turning was evident in the results. Jing, Xin and Lu (2004) analysed the hydrodynamic of C-start. They found that three stages were contained in the C-start process and the change of direction mainly relied on the flap of caudal fin. Epps and Techet (2010) illuminated the formation of two distinct vortices during the turn by using fully time-resolved (500 Hz) PIV. Results showed that the fish body rotation was facilitated by the initial, or "maneuvering" vortex formation, and the final fish velocity was augmented by the strength of the second, "propulsive" vortex. Blake and Chan (2006) presented the relationship between mass and radius using an empirical model. Accordingly, turning radius scales to the 0.37 power of mass. Based on the kinematic data from experimental observation, Yang, Wu and Tong (2009) numerically simulated the routine turns of a Koi carp and found out two kinds of basic turning mode, i.e., single-beat turn and cruising turn. It was noted that single-beat turns were more efficient. Xin and Wu (2018) analysed the rotational moment in favor of turning motion. Results showed that the 3D bionic fish can turn quickly by primarily using the directional control strategy of the head. Wang and Yu (2019) established models of zebrafish's C-turn and C-escape and they found that the duration had a significant influence on escape velocity and energy consumption. Both the C-turn and C-escape have two stages, bending of the body and backward flapping of the tail. The difference is that C-escape has reversed bending while C-turn only flap back to the straight-line state at most. Feng (2020) numerically simulated the 3-DoF C-turn maneuvering of a tuna-like swimmer with the self-propelled motion. The simulations confirmed that two vortex pairs were shed in the turning maneuver of Thunniform swimmers. Based on the kinematic equations of C-turn, Li, Gu and Yao (2021) numerically analyzed the hydrodynamic of the two tunas with erected and depressed median fins during the self-propelled motion. The results illustrated that the morphology of the median fins had a great influence on hydrodynamics and flow field structure during C-turn behaviors.
