在浅海环境中,海底环境参数对声传播有着重要的影响.由于利用单个宽带声源进行海底参数反演时,随着距离的增大,误差变大,本文提出利用warping变换对在浅海波导中传播的,不同距离上的两个宽带爆炸声源进行简正波的有效分离,实现了宽带爆炸声源的远距离海底参数反演.采用全局寻优遗传算法对提取出的模态频散到达时间差与理论计算的模态频散到达时间差进行匹配处理,并结合随距离连续变化的声传播损失,实现了利用单水听器进行海底参数的反演.实验结果表明:运用反演出的海底参数提取模态频散时间差和实测数据提取出的模态频散时间差吻合得较好;而通过传播损失反演得到的海底衰减系数与频率呈指数关系.最后,对反演结果进行了后验概率分析,并将本组爆炸声源的反演结果用于另一组不同距离上爆炸声源时仍然有效,来评价反演结果的有效性.
Acoustic propagation in shallow water is greatly influenced by the properties of the sea bottom. The dispersion characteristics of modes are relatively sensitive to the bottom parameters and have been used to invert the bottom parameters. Since the inversion error using a single wideband sound source increases with increasing range, a far distance inversion method based on the modal dispersion curve using a single hydrophone with two wideband sound sources is presented in this paper, in which a warping transform is applied so that it can accurately extract the modal dispersion curve from the warped signal spectrum. Experimental data used for the inversion are acquired using a hydrophone of vertical array in the South Sea of China during the Autumn in 2012. The transmitted signals are explosive signals, and the bottom sound speed and density are inverted by matching the theoretical arrival time differences of various modes and frequencies with those calculated using the experimental data. The attenuation coefficient is deduced using the transmission loss data recorded in the experiment. A genetic algorithm (GA) is used for optimization search for the parameter bounds. Inversion results demonstrate that the arrival time difference when using the bottom sound speed and density show a high consistency with those obtained using the experimental data. Moreover, the attenuation coefficient is nonlinear over the frequency band from 100 to 315 Hz. The validity of inverted parameters is evaluated by the posteriori probability distributions, and the numerical results of arrival time differences calculated using the inverted sound speed and density are in good agreement with those extracted from the other two wideband explosive signals at different distances. In addition, the theoretical transmission loss calculated using the inverted attenuation coefficient matches the experiment data very well. It is shown that the inversion scheme can provide a valid and stable environmental estimation.