王瑞涵, 陈辉, 管聪. 基于机器学习的船舶机舱设备状态监测方法[J]. 中国舰船研究, 2021, 16(1): 158–166, 192. doi: 10.19693/j.issn.1673-3185.02150
引用本文: 王瑞涵, 陈辉, 管聪. 基于机器学习的船舶机舱设备状态监测方法[J]. 中国舰船研究, 2021, 16(1): 158–166, 192. doi: 10.19693/j.issn.1673-3185.02150
WANG R H, CHEN H, GUAN C. Condition monitoring method for marine engine room equipment based on machine learning[J]. Chinese Journal of Ship Research, 2021, 16(1): 158–166, 192. doi: 10.19693/j.issn.1673-3185.02150
Citation: WANG R H, CHEN H, GUAN C. Condition monitoring method for marine engine room equipment based on machine learning[J]. Chinese Journal of Ship Research, 2021, 16(1): 158–166, 192. doi: 10.19693/j.issn.1673-3185.02150

基于机器学习的船舶机舱设备状态监测方法

Condition monitoring method for marine engine room equipment based on machine learning

  • 摘要:
      目的  为实现船舶机舱设备的智能状态监测,引入机器学习算法,提出一种结合流形学习和孤立森林的船舶机舱设备状态监测方法。
      方法  由于船舶机舱设备的状态监测数据是多维度数据,基于该监测系统,通过流形学习来提取有效的数据特征,实现对原始数据的降维,减少数据复杂度。基于孤立森林算法,在仅利用正常工况数据集的情况下,训练并构建多个子森林检测器,用于实现对目标设备的故障监测。在Matlab/Simulink环境下建立大型船舶二冲程柴油机模型,对其正常工况和故障工况下的数据进行仿真,以验证该方案的有效性。
      结果  通过状态仿真数据对不同故障监测方案性能的比较,验证了所提故障监测方案具有98.5%的故障检测率和3%的故障虚警率。
      结论  所提方法能显著提高船舶机舱设备的故障监测性能。

     

    Abstract:
      Objectives  In order to realize the intelligent condition monitoring of marine engine room equipment, machine learning algorithms are introduced and a condition monitoring method based on manifold learning and an isolation forest is proposed.
      Methods  As condition-monitoring data is multi-dimensional, the proposed method extracts useful features through manifold learning, thereby reducing the dimensions and complexity of the raw data. An isolation forest algrithm is introduced to utilize the normal condition data to train and construct multiple sub forest detectors, realizing the fault monitoring of the target equipment. To validate the proposed scheme, a two-stroke marine diesel engine was developed in Matlab/Simulink to simulate reliable normal and fault condition datasets.
      Results  Comparisons of the simulated datasets of the different fault monitoring schemes demonstrate that the proposed method has a highest fault detection rate of 98.5% and lowest false alarm rate of 3%.
      Conclusions  The method proposed in this study improves the fault monitoring performance of marine equipment.

     

/

返回文章
返回