基于动态学习策略多群体粒子群的消磁站水下磁传感器位置校正方法

doi:10.12382/bgxb.2021.0740

摘要/Abstract

摘要：

消磁站海底敷设磁传感器是舰艇磁场测量的主要形式之一,水下磁传感器的位置偏差直接影响着舰艇磁场的测量精度和防护能力评估。针对现有方法难以准确定位消磁站水下磁传感器的问题,提出一种基于动态学习策略多群体粒子群的消磁站水下磁传感器位置校正方法。该方法首先将通电载流线圈等效成磁偶极子磁源,再通过线性多重计量方法改变磁源与磁传感器的相对位置以获取多组磁传感器磁场测量数据,据此建立水下磁传感器位置校正模型,并采用动态学习策略多群体粒子群优化算法优化求得位置偏差矢量,从而实现水下磁传感器位置的高精度校正。在综合分析磁偶极子等效误差等主要影响因素的基础上,设计了数值模拟实验和物理缩比模型实验,结果表明:该方法可有效解决消磁站水下磁传感器的位置校正问题,校正后x轴、y轴和z轴三个方向的位置误差均小于0.1m。经过校正后的消磁站磁场测量精度可以满足舰艇磁场测量要求,该方法可以对消磁站位置安装偏差不大于0.3m的水下磁传感器完成校正工作,具有较好的实用价值。

关键词: 水下磁传感器, 位置校正, 磁偶极子, 多群体粒子群

Abstract:

Laying magnetic sensors on the seabed of the magnetic deperming facility is one of the main ways to measure ships' magnetic field. The position deviation of underwater magnetic sensors directly affects the measurement accuracy and evaluation of protection ability of ships' magnetic field. To address the problem that the existing methods are difficult to accurately locate the underwater magnetic sensors of the deperming station, a position correction method of underwater magnetic sensors of the deperming station based on multi-population particle swarm optimization (PSO) using dynamic learning strategies is proposed. Firstly, the energized current-carrying coil is equivalent to a magnetic dipole magnetic source, and then the relative position between the magnetic source and the magnetic sensor is changed by the linear multi-measurement method to obtain the magnetic field measurement data of the multiple groups of magnetic sensors. Based on this, the position correction model of underwater magnetic sensors is established, and the position deviation vector is optimized by the multi-population PSO algorithm using dynamic learning strategies, thus realizing the high-precision correction of the position of the underwater magnetic sensors. Based on the comprehensive analysis of the main influencing factors such as the equivalent error of the magnetic dipole, the numerical simulation and physical scale model experiment are designed. The results show that: this method can effectively solve the problem of position correction of underwater magnetic sensors in the deperming station; after correction, the position errors in x,y and z directions are less than 0.1m; the magnetic field measurement accuracy of the deperming station after correction can meet the requirements of ships' magnetic field measurement. This method can complete the correction of the underwater magnetic sensors with a position deviation of no more than 0.3m, and has good practical value.

Key words: underwater magnetic sensors, position correction, magnetic dipole, multi-population particle swarm optimization

中图分类号:

P229

王玉芬, 周国华, 吴轲娜, 李林锋. 基于动态学习策略多群体粒子群的消磁站水下磁传感器位置校正方法[J]. 兵工学报, 2023, 44(2): 526-533.

WANG Yufen, ZHOU Guohua, WU Kena, LI Linfeng. Underwater Magnetic Sensor Position Correction Method Based on Multi-Population Particle Swarm Optimization Using Dynamic Learning Strategies[J]. Acta Armamentarii, 2023, 44(2): 526-533.

图/表 14

图1 潜艇物理模型几何示意图

Fig.1 Geometric diagram of a submarine's physical model

图2 龙骨下磁场对比

Fig.2 Comparison of magnetic field under the keel

表1 磁传感器位置偏差仿真结果

Table 1 Simulation results of sensor position deviation

序号	位置偏差幅值/m	最大绝对误差/nT
1	0.1	43
2	0.2	59
3	0.3	87
4	0.4	100

图3 位置校正原理示意图

Fig.3 Schematic diagram of position correction

图4 载流线圈磁场计算示意图

Fig.4 Diagram of magnetic field calculation of current-carrying coil

图5 动态学习机制方案

Fig.5 Dynamic learning mechanism scheme

图6 仿真模型示意图

Fig.6 Schematic diagram of simulation model

图7 磁偶极子等效误差分析

Fig.7 Equivalent error analysis of magnetic dipole

图8 磁传感器测量精度误差分析

Fig.8 Analysis of sensor measurement accuracy error

图9 环境噪声误差分析

Fig.9 Analysis of ambient noise error in sensor measurement

表2 磁传感器位置校正仿真结果

Table 2 Simulation results of sensor position correction

序号	偏差矢量/m	x轴方向/m	y轴方向/m	z轴方向/m
1	0.3,0,0	0.016	0.009	0.037
2	0,0.3,0	0.020	0.009	0.024
3	0,0,0.3	0.021	0.029	0.021
4	0.3,0.3,0.3	0.022	0.011	0.057
5	-0.3,0.3,0.3	0.021	0.011	0.058
6	0.3,-0.3,0.3	0.016	0.019	0.055
7	0.3,0.3,-0.3	0.018	0.009	0.044
8	-0.3,-0.3,-0.3	0.010	0.007	0.027

图10 模型实验布置图

Fig.10 Layout of model experiment

图11 实验结果

Fig.11 Experimental results

表3 模型实验结果

Table 3 Model's experimental results

序号	偏差矢量/m	x轴方向/m	y轴方向/m	z轴方向/m
1	0.1, 0.1, 0.1	0.0032	0.0044	0.0090
2	-0.1, 0.1, 0.1	0.0051	0.0010	0.0064
3	0.1, -0.1, 0.1	0.0034	0.0081	0.0104
4	0.1, 0.1, -0.1	0.0037	0.0034	0.0053
5	-0.1, -0.1, -0.1	0.0078	0.0069	0.0116

参考文献 17

[1]	王锴松, 周国华, 刘月林, 等. 地磁模拟法测量舰船感应磁场的数值模拟[J]. 兵工学报, 2022, 43(3):617-625. doi: 10.12382/bgxb.2021.0110
	WANG K S, ZHOU G H, LIU Y L, et al. Numerical simulation of measuring ship's induced magnetic field by geomagnetic field simulation method[J]. Acta Armamentarii, 2022, 43(3):617-625. (in Chinese) doi: 10.12382/bgxb.2021.0110
[2]	郭成豹, 胡松, 王文井, 等. 利用磁传感器阵列磁场差值的舰船磁场反演建模方法[J]. 兵工学报, 2022, 43(1):111-119. doi: 10.3969/j.issn.1000-1093.2022.01.012
	GUO C B, HU S, WANG W J, et al. Ship magnetic field inversion modeling method utilizing the magnetic field difference between magnetic sensors[J]. Acta Armamentarii, 2022, 43(1):111-119. (in Chinese) doi: 10.3969/j.issn.1000-1093.2022.01.012
[3]	HOTT M, HARLAKIN A, HOEHER P A. Hybrid communication and localization underwater network nodes based on magnetic induction and visible light for AUV support[C]// Proceedings of 2020 International Conference on Information and Communication Technology Convergence.Jeju, Korea:IEEE, 2020:66-68.
[4]	HUANG Y, WAN B W, WU L H. Theoretical investigation on the linear location algorithm of the magnetic gradient tensor ranging by use of cuboid triaxial magnetometer array[J]. IEEE Transactions on Magnetics, 2021, 57(11):1-11.
[5]	AMIR L, ELAD F, ROGER A, et al. Multiclass permanent magnets superstructure for indoor localization using artificial intelligence[J]. IEEE Transactions on Magnetics, 2021:1-1.
[6]	PASKU V, ANGELIS A D, ANGELIS G D, et al. Magnetic field-based positioning systems[J]. IEEE Communications Surveys & Tutorials, 2017, 19(3):2013-2017.
[7]	CALLMER J, SKOGLUND M, GUSTAFSSON F. Silent localization of underwater sensors using magnetometers[J]. Eurasip Journal on Advances in Signal Processing, 2010, 2010(1):709318. doi: 10.1155/2010/709318 URL
[8]	张朝阳, 肖昌汉. 海底布放磁传感器的磁定位方法的模拟实验研究[J]. 上海交通大学学报, 2011, 45(6):826-830.
	ZHANG C Y, XIAO C H. Simulation experiment research for magnetic localization method for magnetometer sensor at seabed[J]. Journal of Shanghai Jiao Tong University, 2011, 45(6):826-830. (in Chinese)
[9]	杨明明, 刘大明, 连丽婷, 等. 用海面磁偶极子源定位海底矢量磁传感器[J]. 探测与控制学报, 2011, 33(5):41-45.
	YANG M M, LIU D M, LIAN L T, et al. Underwater vector-magnetometer located by using magnetic dipole sources in the offing[J]. Journal of Detection & Control, 2011, 33(5):41-45. (in Chinese)
[10]	孙晔, 肖昌汉, 周国华. 磁主动式海底磁传感器定位方法及其解析公式[J]. 海洋测绘, 2012, 32(4):25-32.
	SUN Y, XIAO C H, ZHOU G H. An active method and its analytical formulas of magnetic sensor positioning using the magnetic source[J]. Hydrographic Surveying & Charting, 2012, 32(4):25-32. (in Chinese)
[11]	周国华, 张树, 赵文春, 等. 低速运动条件下的磁传感器定位方法[J]. 船电技术, 2016, 36(9):1-5.
	ZHOU G H, ZHANG S, ZHAO W C, et al. Magnetic sensor positioning method under low speed movement condition[J]. Marine Electric & Electronic Engineering, 2016, 36(9):1-5. (in Chinese)
[12]	BIAN S Z, HEVESI P, CHRISTENSEN L, et al. Induced magnetic field-based indoor positioning system for underwater environments[J]. Sensors, 2021, 21(6):2218. doi: 10.3390/s21062218 URL
[13]	NAUREEN A, ZHANG N, FURBER S, et al. A gps-less localization and mobility modelling (LMM) system for wildlife tracking[J]. IEEE Access, 2020, 8:102709-102732. doi: 10.1109/Access.6287639 URL
[14]	ZHOU Y, XIAO C H, ZHOU G H. Multi-objectivization-based localization of underwater sensors using magnetometers[J]. IEEE Sensors Journal, 2014, 14(4):1099-1106. doi: 10.1109/JSEN.2013.2287915 URL
[15]	张新明, 王霞, 涂强, 等. 融合榜样学习和反向学习的粒子群优化算法[J]. 河南师范大学学报(自然科学版), 2017, 45(6):91-99.
	ZHANG X M, WANG X, TU Q, et al. Particle swarm optimization algorithm based on combing example learning and opposition learning[J]. Journal of Henan Normal University (Natural Science Edition), 2017, 45(6):91-99. (in Chinese)
[16]	YE W X, FENG W Y, FAN S H. A novel multi-swarm particle swarm optimization with dynamic learning strategy[J]. Applied Soft Computing, 2017, 61:832-843. doi: 10.1016/j.asoc.2017.08.051 URL
[17]	WU J S, CHEN E H. A novel hybrid particle swarm optimization for feature selection and kernel optimization in support vector regression[C]// Proceedings of 2010 International Conference on Computational Intelligence and Security.Nanning, China:IEEE, 2010:189-194.