基于声信号海温垂直结构反演试验  

Inversion experimentation of vertical ocean temperature structure based on acoustical signal

作　　者：毛科峰[1] 陈希[1] 李妍[1,2] 李振锋 梁毅[1] 

机构地区：[1]解放军理工大学气象学院 [2]东南大学自动化学院 [3]解放军63898部队

出　　处：《解放军理工大学学报（自然科学版）》2011年第6期690-695,共6页Journal of PLA University of Science and Technology(Natural Science Edition)

基　　金：卫星海洋环境动力学国家重点实验室开放研究基金资助项目(SOED1009)

摘　　要：目前利用声回波记录反演海洋温度垂直结构的方法中大多以声阻抗剖面作为反演对象,不便于实现声速和密度分离。针对此问题,提出了一种新的反演思路。在优化海温垂直结构参数化表达的基础上,利用遗传算法进行参数寻优,建立了海温垂直结构反演模型,以海温垂直剖面实测数据检验了模型的准确性。结果表明:反演模型能够直接得到温盐垂直结构反演结果,对混合相位信号的回波进行反演时,精度较高,且不需要对信号进行最小相位化处理,较之层剥算法更简便。Currently most methods for the inversion of the vertical ocean temperature structure with raw acoustical echo use acoustic impedance profile as the inversion object,and can not separate velocity of sound and density easily,a new inversion idea was presented making use of the method of optimization parameter and builting inversion model based on optimizing the parameter expression of the vertical ocean temperature profile to verify model＇s correctness by actual measure data of the vertical ocean temperature structure.The result shows that this inversion model can get the inversion result of the vertical ocean temperature structure directly which has a high precision compared to the result inversed by the mix-phase signal＇s echo,and the model does not need the smallest phase signal,so it＇s more simple and convenient than Layered Peeling algorithm.

关 键 词：海温垂直结构 混合相位信号 遗传算法 

分 类 号：P731[天文地球—海洋科学]