Feasibility of Wave Measurement by Using a Sailing Buoy and the Artificial Neural Network Technique

Abstract

Ocean waves have adverse effects on the operation and safety of ships. Real-time and accurate onboard wave monitoring is essential for green smart ships and navigational safety. It is an important method to use ship motion to perform inversion of the information of encountering waves, which has many advantages such as low cost and high spatio-temporal resolution, and has attracted wide attention. An inversion model based on artificial neural network (ANN) is proposed to deal with the shortcomings of existing ship motion inversion ocean wave models. Taking the motion time history of the ship as the input and the wave elevation time history as the output, the artificial neural network is used to extract the motion features of the ship, and the linear function is input to perform wave surface time history inversion. To verify the feasibility and the accuracy of the inversion model, a ship model tank test is conducted. The results show that the proposed wave measurement method based on the artificial neural network can well achieve regular wave and irregular wave elevation time history inversion. Most of the statistical errors of regular wave inversion are less than 10% and the error of irregular wave inversion is about 10%. The method proposed provides an effective and feasible means for inversion of wave time information in ship motion.

Keywords

• wave inversion
• artificial neural network
• sailing wave buoy
• tank testing