基于BP神经网络的磁异常向下延拓方法

Downward extension method of magnetic anomaly based on BP neural network

舰船、潜艇等水中目标会产生磁异常场,这一物理场中包含丰富的目标固有信息。航空磁测是一种快速高效的磁测方式,通过一定的向下延拓算法,能够准确识别定位目标。但现有的磁异常向下延拓方法延拓深度有限,不能从航磁数据向下延拓至海面。BP神经网络方法具有较强的非线性映射能力,在函数逼近和模式识别等领域具有广泛的应用。本文以BP神经网络方法为基础,建立一种磁异常数据向下延拓方法,经理论模型验证,本方法准确可靠,能够较稳定地进行30倍点距的向下延拓,达到从航磁数据延拓至海面的目的。

Underwater targets such as ships and submarines will generate magnetic anomalous fields,which contain rich target inherent information.Aerial magnetic survey is a fast and efficient magnetic survey method,which can accurately identify and locate the target through a certain downward extension algorithm.However,the existing downward extension methods of magnetic anomalies have limited extension depth and cannot extend downward from aeromagnetic data to the sea surface.The BP neural network method has strong nonlinear mapping ability,and has a wide range of applications in the fields of function approximation and pattern recognition.Based on the BP neural network method,this paper establishes a downward continuation method of magnetic anomaly data.The theoretical model has verified that this method is accurate and reliable,and can stably carry out downward continuation of 30 times the point distance,so as to achieve the goal of continual navigation.The purpose of extending magnetic data to the sea surface.