Ducted propeller is a kind of special propeller widely used in unmanned underwater vehicles, its flow characteristics and hydrodynamic noise are very important for marine environmental protection and equipment concealment. The hybrid techniques based on the acoustic analogy theory are adopted in the present study to calculate the unsteady flow field and sound field characteristics of a ducted propeller. The full scale flow filed and hydro-acoustic sources of the propulsion system are simulated by Detached-Eddy computational fluid dynamics (CFD) method. Hydrodynamic noise are calculated by FWH equation based on the CFD results. The frequency domain and directivity of sound pressure level at different sound field monitoring points are analyzed at four navigational speeds. The results show that the navigational speed that is in the inflow condition of the ducted propeller play important roles in the flow structure and underwater radiated noise. Under the fixed impeller rotational speed, the propulsion efficiency of ducted propeller increases first and then decreases with the raise of navigational speed. The maximum errors of thrust and power between simulation and experiment values are 0.5% and 0.1% respectively, which means that the adopted DES numerical simulation method has high credibility in calculating the acoustic source. At impeller rotational speed of 2000 r/m, the best state of flow field distribution is at the navigational speed of 1.54 m/s, which is corresponding to the highest propulsion efficiency condition. The propeller noise presents dipole characteristic in all working conditions, and at the obvious blade passing frequency, multiple characteristics are presented; most of the noise contribution is also concentrated below four times of the blade passing frequency. The total sound pressure level of the hydrodynamic noise is the smallest at the optimal efficiency condition (the navigational speed is 1.54 m/s). At high navigational speed, the low frequency characteristics below blade passing frequency increase and the amplitude becomes larger. This indicates that the component of turbulent noise becomes more important with the increase of navigational speed. The research focuses on analyzing the relationship between the energy loss of the ducted propeller wake field and the noise level, and it is found that the vortex at the tail makes a certain contribution to the noise. The research conclusions could provide some reference for the acoustic performance evaluation and noise reduction optimization of ducted propeller design as well as the improvement of UUV stealth performance.

An innovative numerical prediction method of flow-induced noise is implemented to overcome the defect that the traditional acoustic analogy method cannot reflect the interaction between turbulence vortex and sound. The classical cylindrical–airfoil interference model is used to perform the simulation and compared with the experimental results. To start with the derivation of Powell’s vortex sound equation, an implicit three-dimensional model of the fluid–acoustic coupling field is established to process the unsteady iterative calculation. The large eddy simulation method is adopted to solve the unsteady flow, and the acoustic information is then calculated using the vortex acoustic equation at each iteration step. The vortex structures around the cylinder airfoil are identified and captured by the Q-criterion for further analysis of vortex–noise correlation mechanism. The flow-induced noise prediction results are finally compared with Ffowcs Williams–Hawkings (FW–H) acoustic analogy approach. The results show that the vortex shedding from the cylinder and the interaction between vortex shedding and airfoil have the greatest influence on the acoustic, and the far-field noise of the cylinder airfoil shows a partial “eight” dipole distribution. The calculated results of the vortex sound theory are closer to the experimental ones than the FW–H method. The research helps understand the vortex acoustic coupling mechanism of the cylinder–airfoil model and provides a more accurate numerical prediction of flow-induced noise.

Most of the research on the cavitation with entrained air has focused on the military direction, but it ,about centrifugal pumps, which is relevant to people's livelihood, is still relatively lacking. In order to study the basic law of the development of cavitation inside centrifugal pumps under aeration conditions, a test bench suitable for cavitation experiments with incoming flow containing gas was obtained. Furthermore, a single-stage single-suction 6-blade centrifugal pump was used as the research object to conduct pressure pulsation experiments under cavitation condition when the incoming flow was 1.0% air viod fraction at 2900r/min-50m3/h. The results showed that: After cavitation happened, the greater aeration content will deteriorate the pump's anti-cavitation performance, but the head curve is more gentle in falling down compared to natural cavitation. Hence aeration has a beneficial effect on the performance degradation of the pump under the cavitation condition. At the same time, before the cavitation number drops to the fracture cavitation number of the pump, aeration has improvement in the efficiency of the pump in different degrees , especially in the situation with the ventilated rate of 1.0%. The main frequency of pressure pulsation at the inlet and outlet of the test pump after aeration is dominated by the blade frequency. The shaft frequency signal at the outlet gradually decreases with the cavitation number lessened. Moreover the amplitude of the blade frequency grows slightly with the reduction of the cavitation number. But it tends to soar when the cavitation number is less than the fracture cavitation number.