Measurement distance of underwater elastic cylindrical shell for radiated sound power

For underwater acoustic measurement experiment, if the scale of underwater elastic structure is large, it is difficult to achieve the requirement of the classic far-field testing criterion. This paper has researched the measuring distance criterion for underwater structure radiated noise from the perspective of energy measurement. A concept of measuring distance criteria is derived responding to the far-field criterion of traditional acoustic theory. A numerical simulation is presented for some finite elastic cylindrical shells excited by a point force in the water. The radiation sound field at points on the acoustic axis of the shell is simulated by the finite element method (FEM) and boundary element method (BEM). The pressure and velocity are numerical calculated at the annular field points in the three coordinate planes. The directivity and the sound pressure amplitude attenuation characteristic on the field points are analyzed to research the most suitable distance for measuring the radiated sound power. Then, a calculation method for the measuring distance has been preliminarily summarized. The simulation results show that the radiated sound power from the cylindrical shell can be relatively precisely measured at the distance determined by the numerical simulation results. And this distance is much less than the classic criterion of far-field for sound sources in general. The experimental data of an underwater cylindrical shell is processed and analyzed. The experimental results demonstrate that: the attenuation law of sound pressure amplitude at the criterion distance determined by the calculation method proposed in this paper is very similar to the attenuation law of spherical waves. This means that if the radiated sound power is calculated by the mean squared sound pressure method, the measurement distance could be nearer to the underwater elastic structure. It is very valuable for sound source level measurements when the signal-to-noise ratio (SNR) is low. And more, because the phase does not need to be measured, the problem caused by the poor accuracy of phase measurement can be avoided.