﻿ 基于组合优化算法的船舶信息风险评估
 舰船科学技术  2024, Vol. 46 Issue (5): 163-166    DOI: 10.3404/j.issn.1672-7649.2024.05.030 PDF

Ship information risk assessment based on combinatorial optimization algorithm
ZHU Guo-jun
Marine College, Zhejiang Institute of Communications, Hangzhou 311112, China
Abstract: In order to avoid major ship navigation accidents caused by ship information risks, a ship information risk assessment method based on combinatorial optimization algorithms is studied. Selecting a total of 17 indicators from four aspects of communication, environment, management, and human factors, a ship information risk assessment index system is constructed. It is used as input data for the radial basis function (RBF) neural network input layer, and after hidden layer mapping operation, the evaluated ship information risk level is output through the output layer. A combination optimization algorithm combining fuzzy C-means clustering algorithm and genetic algorithm is adopted, Reasonably selecting the center vector of the hidden layer in the RBF neural network and optimizing it to obtain the optimal width and weight vector of the hidden layer basis function, in order to improve the effectiveness of ship information risk assessment. The experimental results show that this method can effectively evaluate the information risk of multiple ships, and based on the evaluation results, identify the factors that cause ship information risk and provide targeted guidance suggestions.
Key words: combinatorial optimization     ship information     risk assessment     indicator system     RBF neural network     genetic algorithm
0 引　言

1 基于组合优化算法的船舶信息风险评估 1.1 船舶信息风险指标体系构建

 图 1 船舶信息风险评估指标体系 Fig. 1 Indicator system for ship information risk assessment

1.2 组合优化算法改进RBF的船舶信息风险评估 1.2.1 基于RBF神经网络的船舶信息风险评估

 图 2 RBF神经网络结构 Fig. 2 RBF neural network structure
 $\varphi ({x_i},{v_k}) = \exp \left( { - \frac{1}{{2\sigma _k^2}}{{\left\| {{x_i} - {v_k}} \right\|}^2}} \right)。$ (1)

 ${y_i} = \sum\limits_{k = 1}^S {{w_{ik}}} \varphi ({x_i},{v_k}) + {\varepsilon _i} 。$ (2)

 $J(U,V) = \sum\limits_{i = 1}^c {\sum\limits_{j = 1}^n {u_{ij}^md_{ij}^2} }。$ (3)

FCM实现中心向量选取的步骤如下：

 ${{{U}}^{(g + 1)}}={u_{ij}}{\left[ {\sum\limits_{r = 1}^c {{{\left( {\frac{{{d_{ij}}}}{{{d_{rj}}}}} \right)}^{\frac{2}{{m - 1}}}}} } \right]^{ - 1}}，$ (4)

 ${V^{(g + 1)}}={v_i}\frac{{\sum\limits_{j = 1}^n {u_{ij}^m{x_j}} }}{{\sum\limits_{j = 1}^n {u_{ij}^m} }} 。$ (5)

2）基于遗传算法的参数优化流程

2 结果与分析

 图 3 6艘船舶信息风险评估结果 Fig. 3 Results of information risk assessment for 6 ships

 图 4 船舶6的各指标风险评估结果 Fig. 4 Risk assessment results of various indicators for ship 6
3 结　语

 [1] 江忠. 船舶信息分析系统风险评价数学模型研究[J]. 舰船科学技术, 2020, 42(16): 49-51. JIANG Zhong. Research on mathematical model for risk assessment of ship information analysis system[J]. Ship Science and Technology, 2020, 42(16): 49-51. [2] 陈毕伍, 沙正荣, 吴建生, 等. 基于船舶领域的珠江口通航风险评估[J]. 大连海事大学学报, 2020, 46(1): 29-38. CHEN Bi-wu , SHA Zheng-rong , WU Jian-sheng , et al. Navigation risk assessment of Pearl River Estuary based on ship domain[J]. Journal of Dalian Maritime University, 2020, 46(1): 29-38. [3] 詹锦皓, 李维波, 李齐, 等. 基于比例伪时序算法的舰船电力风险评估系统[J]. 中国舰船研究, 2022, 17(1): 176-186. ZHAN Jin-hao , LI Wei-bo , LI Qi , et al. Ship power risk assessment system based on proportional pseudo time-series algorithm[J]. Chinese Journal of Ship Research, 2022, 17(1): 176-186. [4] 薛彦卓, 周莹, 鲁阳, 等. 基于模糊AHP-DEMATEL的北极冰区船舶冰困风险评价[J]. 哈尔滨工程大学学报, 2022, 43(7): 944-949+992. XUE Yan-zhuo, ZHOU Ying, LU Yang, et al. Risk assessment of ships stuck in ice in arctic ice area based on fuzzy AHP-DEMATEL[J]. Journal of Harbin Engineering University, 2022, 43(7): 944-949+992. [5] 范中洲, 郭婷婷, 郑力铭. 基于改进集对分析法的船舶碰撞风险评价[J]. 安全与环境学报, 2021, 21(2): 470-474. FAN Zhong-zhou, GUO Ting-ting, ZHENG Li-ming. Assessment on the ship collision risk based on the improved set pair analysis method[J]. Journal of Safety and Environment, 2021, 21(2): 470-474. [6] 杜沛, 曾喆, 任利锋, 等. 基于动态海洋环境要素的航行风险评估方法研究[J]. 海洋科学, 2021, 45(5): 121-129. DU Pei , ZENG Zhe , REN Li-feng , et al. Navigation risk assessment method based on dynamic marine environmental factors[J]. Marine Sciences, 2021, 45(5): 121-129.