基于EMI与ANN技术水下固定频率信号探测研究

发布时间：2018-08-16 13:51

【摘要】：随着海洋技术与海运业的发展,针对探寻水下信号的需求愈加迫切。无论是对于探测失事落水后的飞机黑匣子,还是寻找其它在水中发射固定频率信号构件,能否实现快速的探测和定位,对保护人身安全与社会财产都有着重要的意义。本文针对目前水下信号探测情况复杂,对信号检测影响因素较多的情况为背景,基于压电阻抗技术设计了水下固定频率信号构件的实验装置及探测理论,搭建了模拟超声信标探测实验平台,并在水下进行探测实验,最终结合BP神经网络技术与三球定位原理实现对目标固定频率的定位。本文主要研究内容归纳如下:(1)综述了国内外目前对水下频率信号探测识别的研究现状,介绍了压电材料和EMI(压电阻抗技术)技术的应用研究现状,以及未来发展的方向,介绍了 ANN(人工神经网络)技术对科学研究中的应用,阐述了压电耦合电阻抗理论,并对压电阻抗与机械阻抗之间的联系作了分析。(2)利用ANSYSAPDL中压电分析模块对压电陶瓷片进行有限元分析,分析其变形量以及主要变形方向,对半球形压电阻抗接收器基座进行自由模态分析,得到其在目标信号附近处大致固有频率,为实验提供一定的参考。(3)设计了基于压电阻抗技术的固定频率信号探测装置,搭建了模拟探测固频信号的实验平台,发现了目标频率信号与扫频信号同频共振现象,对于不同位置,不同潜水深度下进行两组不同固频信号探测研究。(4)提出了神经网络定位与三球定位的基本方法理论,创建了用于定位目标固频信号的BP神经网络,使用训练完成后的网络对目标信号进行定位研究,结合三球定位原理,证明了利用神经网络技术于压电阻抗技术相结合对目标固频信号探测及定位的可行性。
[Abstract]:With the development of ocean technology and shipping industry, it is more and more urgent to explore the underwater signal. Whether it is for detecting the aircraft black box after the crash or looking for other components that transmit fixed frequency signals in the water, it is of great significance to protect personal safety and social property. Aiming at the complicated underwater signal detection and the situation that there are many factors affecting the signal detection, this paper designs the experimental device and detection theory of the underwater fixed frequency signal component based on piezoelectric impedance technology, builds the simulation ultrasonic beacon detection experimental platform, and carries on the detection experiment under water, finally combines with BP neural network. The main contents of this paper are summarized as follows: (1) The research status of underwater frequency signal detection and recognition at home and abroad is summarized, and the application of piezoelectric materials and EMI (piezoelectric impedance technology) technology is introduced, as well as the future development direction, and ANN (human) is introduced. The theory of piezoelectric coupling impedance is expounded and the relationship between piezoelectric impedance and mechanical impedance is analyzed. (2) Finite element analysis of piezoelectric ceramic sheet is carried out by using ANSYS APDL medium piezoelectric analysis module, and the deformation amount and main deformation direction are analyzed. Free modal analysis is carried out on the base of the receiver, and the approximate natural frequency near the target signal is obtained, which provides a certain reference for the experiment. (3) A fixed frequency signal detection device based on piezoelectric impedance technology is designed, and an experimental platform is built to simulate the detection of fixed frequency signal. (4) The basic theory of neural network positioning and three-ball positioning is put forward, and the BP neural network used to locate the fixed-frequency signal of the target is created. After training, the network is used to locate the target signal. It is proved that it is feasible to detect and locate the fixed frequency signal of the target by combining the neural network technique with the piezoelectric impedance technique.
【学位授予单位】：安徽理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P715.5

【相似文献】