Research On Real-time Measurement Method Of Hull Deformation Angle

Hull deformation Angle is the main reason for the ship space coordinate system can not be established the benchmark,in order to provide the ship of the high precision attitude reference and improve the transfer alignment accuracy,must eliminate the influence of hull deformation and forecast in the hull deformation Angle,how to accurately measure the hull deformation Angle and the local coordinate system to make use of the unified coordination is the essential means to improve the efficiency of the shipboard equipment work together.In this paper,several common methods of hull deformation Angle estimation are introduced.Then,a propagation neural network algorithm based on BP(Back Propagation)neural network is proposed.The main work can be summarized as follows:1.This paper introduces the main causes and classification of hull deformation Angle,studies the current measurement technology of hull deformation Angle at home and abroad,and analyzes the advantages and disadvantages of various methods.The Kalman filter based on angular velocity matching is introduced in detail,including filter design,hull deformation Angle modeling,etc.The existing problems of this method are pointed out and the solutions are proposed.2.The particle swarm optimization algorithm was improved and combined with the Kalman filter based on angular velocity matching to estimate the hull deformation Angle,so as to identify the parameters of the second-order Markov model of the hull deformation Angle.Through analysis and experiment,it is proved that the improved PSO has better identification effect than the standard PSO.3.The basic principle of BP neural network to estimate hull deformation Angle is introduced in detail,the influence of various variables in the neural network on hull deformation estimation is analyzed,and a system of BP neural network to estimate hull deformation Angle is built.The angular velocity data output by the gyroscope and the real hull deformation Angle are collected to train the BP neural network.The effectiveness of the BP neural network model for estimating hull deformation Angle is verified by simulation.4.The improved particle swarm optimization algorithm was combined with BP neural network model to estimate hull deformation Angle,and the weights and thresholds of the neural network were optimized.The simulation verified that the BP neural network optimized by the improved particle swarm optimization algorithm had higher estimation accuracy and performance in estimating hull deformation Angle.Based on the comprehensive comparison of several estimation methods of hull deformation Angle used in this paper,it is concluded that the BP neural network optimized by improved particle swarm optimization and the Kalman filter based on angular velocity matching estimate of hull deformation Angle have the same accuracy,the algorithm is simple and easy to realize,and the convergence is faster,with high engineering value.