关键词: 舰船磁场, 磁偶极子阵列分布, 测量数据, 系数矩阵条件数, 模拟退火算法, 逐步回归

Abstract: In modeling ship magnetic field with hybrid model,the dipole array location distribution and the authenticity of the measurement data,which affects model accuracy and stability,should be considered.The paper takes the condition number of the coefficient matrices of the hybrid model as the objective function,gets the optimal position distribution of magnetic dipole arrays by optimizing the matrix condition number with simulated annealing algorithm,optimizes the measurement data by eliminating the great error in the measurement point,and finally obtains the high precision and stable model of ship magnetic field by using stepwise regression to solve the model equation.The measured data model shows,compared with the model based on the equidistant distribution of magnetic dipole of hybrid model,the model that the proposed method builds has a higher accuracy and is more stable,and the accuracy of the model can reach 96%.This method can be used for high precision modeling of ship magnetic field under complex sea conditions,as well as surface magnetic target detection and positioning system.

Key words: ship magnetic field, dipole array location distribution, measurement data, condition number of matrices, simulated annealing algorithm, stepwise regression