﻿ 决策树算法在船舶自主巡航数据消冗中的应用
 舰船科学技术  2024, Vol. 46 Issue (12): 157-161    DOI: 10.3404/j.issn.1672-7649.2024.12.027 PDF

Application of decision tree algorithm in redundancy reduction of ship autonomous cruise data
SHENG Lijun, CHEN Shiqi
Wuhan Institute of Shipbuilding Technology, Wuhan 430050, China
Abstract: Intelligent management and navigation of ships need to be completed according to reliable autonomous cruise data, and a large number of sensor data and monitoring information are used as input, so that the system can make correct decisions. However, these data may have redundant information interference, which affects the reliability of intelligent decision-making system. Therefore, the application of decision tree algorithm in ship autonomous cruise data redundancy is studied. Filtering, interpolation and hybrid time series data generation are used to process the time series of ship autonomous cruise data and generate standardized time series data of ship autonomous cruise. According to the processed data, a decision tree is generated to classify the autonomous cruise data of the ship; By calculating the data similarity between the same kind and designing the eliminator, the ship autonomous cruise data can be eliminated and the cruise data without redundancy can be obtained. The test results show that the algorithm has a good effect on data time series processing, which can divide different data categories and calculate the similarity between similar data, and the maximum space reduction ratio is 27.8%.
Key words: decision tree algorithm     autonomous cruising of ships     data redundancy elimination     time series data     data similarity     data classification
0 引　言

1 船舶自主巡航数据消冗 1.1 船舶自主巡航数据时序处理

 图 1 船舶自主巡航数据的时序处理流程 Fig. 1 Time series processing flow of ship autonomous cruise data

 $I\left( {{z_1},{z_2},...,{z_k}} \right) = - \sum\limits_{i = 1}^m {p\left( {{{\tilde x}_i}} \right)} \log p\left( {{{\tilde x}_i}} \right)。$ (3)

 $\sum\limits_{i = 1}^m {p\left( {{{\tilde x}_i}} \right) = 1}。$ (4)

 $\psi \left( {X,s} \right) = - \sum\limits_{i = 1}^k {{{\left( {\frac{X}{s}} \right)}_i}}。$ (5)

$\mu \left( {i,j} \right) > G$，表示该数据为冗余数据；当$\mu \left( {i,j} \right) \leqslant G$表示数据为非冗余数据。

2）消除器设计

 图 2 消除器结构 Fig. 2 Structure of eliminator

2 测试分析

 图 3 船舶自主巡航数据的时序处理效果 Fig. 3 Time series processing effect of ship autonomous cruise data

 图 4 船舶自主巡航数据分类结果 Fig. 4 Classification results of ship autonomous cruise data

3 结　语

 [1] 杨晨, 邓茹凤, 张宏, 等. 基于网络通信的设备互操和数据热备份的设计方法[J]. 船海工程, 2022, 15(5): 11-14. YANG Chen, DENG Rufeng, ZHANG Hong, et al. Design method for equipment interoperation and data hot-backup based on network communication[J]. Ship & Ocean Engineering, 2022, 15(5): 11-14. [2] 张泽辉, 管聪, 高航, 等. 面向船联网的高效隐私保护联邦学习方法[J]. 中国舰船研究, 2022, 17(6): 48-58. ZHANG Zehui, GUAN Cong, GAO Hang, et al. Efficient privacy-preserving federated learning method for Internet of Ships[J]. Chinese Journal of Ship Research, 2022, 17(6): 48-58. [3] 陈小娟, 贺红艳, 张慧萍. 基于数据消冗技术的隐私大数据属性加密仿真[J]. 计算机仿真, 2022, 39(11): 422-426. CHEN Xiaojuan, HE Hongyan, ZHANG Huiping. Simulation of privacy big data attribute encryption based on data de redundancy technology[J]. Computer Simulation, 2022, 39(11): 422-426. [4] 高志宇, 宋学坤, 肖俊生, 等. 基于神经网络的大规模数据集离群点检测算法[J]. 沈阳工业大学学报, 2022, 44(4): 420−425. GAO Zhiyu, SONG Xuekun, XIAO Junsheng, et al. Outlier detection algorithm for large-scale data sets based on neural network[J]. Journal of Shenyang University of Technology, 2022, 44(4): 420−425. [5] 李颜平, 吴刚. 基于典型数据集的数据预处理方法对比分析[J]. 沈阳工业大学学报, 2022, 44(2): 185-192. LI Yanping, WU Gang. Comparative analysis of data preprocessing methods based on typical data set[J]. Journal of Shenyang University of Technology, 2022, 44(2): 185-192. [6] 张莉, 丁毛毛, 李玮, 等. 基于决策树算法的客服终端冗余数据迭代消除方法[J]. 计算技术与自动化, 2022, 41(4): 118-122. ZHANG Li, DING Maomao, LI Wei, et al. Iterative elimination method of redundant data in customer service terminal based on decision tree algorithm[J]. Computing Technology and Automation, 2022, 41(4): 118-122. [7] 任兵, 郭艳, 李宁, 等. 基于压缩感知的相关性数据填补方法[J]. 计算机科学, 2023, 50(7): 82-88. REN Bing, GUO Yan, LI Ning, et al. Method for Correlation Data Imputation Based on Compressed Sensing[J]. Computer Science, 2023, 50(7): 82-88. [8] 赵雪, 李晓会. 面向非独立同分布数据的联邦梯度提升决策树[J]. 计算机应用研究, 2023, 40(7): 2184-2191. ZHAO Xue, LI Xiaohui. Federated gradient boosting decision tree for non-IID dataset[J]. Application Research of Computers, 2023, 40(7): 2184-2191. [9] 王泷德, 曹辉, 魏来. 不平衡数据下船舶主机在线故障诊断研究[J]. 中国舰船研究, 2023, 18(5): 269-275. WANG Longde, CAO Hui, WEI Lai. Study on fault diagnosis of marine main engine's online imbalanced data[J]. Chinese Journal of Ship Research, 2023, 18(5): 269-275. [10] 李根, 翟伟, 邬岚. 基于梯度提升决策树的汇合交互作用研究[J]. 浙江大学学报(工学版), 2022, 56(4): 649-655. LI Gen, ZHAI Wei, WU Lan. Study of merging interactions based on gradient boosting decision tree[J]. Journal of Zhejiang University(Engineering Science), 2022, 56(4): 649-655.