上海交通大学学报(英文版) ›› 2015, Vol. 20 ›› Issue (2): 224-227.doi: 10.1007/s12204-015-1611-1

• • 上一篇    下一篇

Neural Network Prediction Model for Ship Hydraulic Pressure Signal Under Wind Wave Background

LI Song1 (李松), ZHANG Chun-hua2* (张春华), SHI Min1 (石敏)   

  1. (1. Naval Academy of Armament, Beijing 100161, China; 2. Unit 61741, Beijing 100094, China)
  • 发布日期:2015-04-02
  • 通讯作者: ZHANG Chun-hua (张春华) E-mail:xiaohua080506@163.com

Neural Network Prediction Model for Ship Hydraulic Pressure Signal Under Wind Wave Background

LI Song1 (李松), ZHANG Chun-hua2* (张春华), SHI Min1 (石敏)   

  1. (1. Naval Academy of Armament, Beijing 100161, China; 2. Unit 61741, Beijing 100094, China)
  • Published:2015-04-02
  • Contact: ZHANG Chun-hua (张春华) E-mail:xiaohua080506@163.com

摘要: The ship hydraulic pressure signal is one of the important characters for the target detection and recognition. At present, most of the researches on the detection focus on the ways in the time domain. The ways are usually invalid in the large wind wave background. In order to solve the problem efficiently, we present an effectual way to detect the ship using the ship hydraulic pressure signal. Firstly, the signature in the proposed method is decomposed by wavelet-transform technique and reconstructed at the low-frequency region. Then, a predictive model is set up by using the radial basis function (RBF) neural network. Finally, the signature predictive error is regarded as the testing signal which can be used to judge whether the target exists or does not. The practical result shows that the method can improve the signal to noise ratio (SNR) obviously.

关键词: hydrodynamic pressure signal, wavelet-transform, radial basis function (RBF), neural network, signal to noise ratio (SNR), predictive error

Abstract: The ship hydraulic pressure signal is one of the important characters for the target detection and recognition. At present, most of the researches on the detection focus on the ways in the time domain. The ways are usually invalid in the large wind wave background. In order to solve the problem efficiently, we present an effectual way to detect the ship using the ship hydraulic pressure signal. Firstly, the signature in the proposed method is decomposed by wavelet-transform technique and reconstructed at the low-frequency region. Then, a predictive model is set up by using the radial basis function (RBF) neural network. Finally, the signature predictive error is regarded as the testing signal which can be used to judge whether the target exists or does not. The practical result shows that the method can improve the signal to noise ratio (SNR) obviously.

Key words: hydrodynamic pressure signal, wavelet-transform, radial basis function (RBF), neural network, signal to noise ratio (SNR), predictive error

中图分类号: