基于小波和SOM网络的船舶管系泄漏检测方法

Ship Piping System Leakage Detecting Based on Wavelet and SOM Network

为了实现对船舶管系的实时泄漏检测,降低负压波法的误报警率,必须将管系泄漏与调泵、调阀等常规操作识别开来。根据不同工况下管道压力和流量信号的波形特征不同,利用小波变换模极大值提取船舶管系工况的特征向量;设计和组建SOM网络(自组织特征映射神经网络),将工况特征向量输入SOM网络,通过网络训练建立输入与输出间良好的非线性映射;利用SOM网络抽取输入信号模式特征的能力,实现了船舶管系泄漏识别。经实验验证,该方法具有良好的准确度和适应性。

In order to monitor ship piping system leakage and reduce false alarm rate of negative pressure wave method,piping leakage must be identified.Because the pipeline pressure and flow signals had different waveform characteristics and singularity,wavelet transform modulus maxima could be used to extract working condition eigenvectors,and then SOM network(self-organizing feature map neural network) was used to identify leakage.From data sampling in terms of pipe working conditions,learning samples were obtained.Accordingly,the nonlinear mapping between SOM neural network inputs and outputs were well established via training.Afterwards,ship piping leakage was detected based on input eigenvectors.The experiments confirm that this method has good accuracy and adaptability.