一种基于k-Means算法的船舶主机工况二次划分方法  

Two-Stage Classification of Main Engine Working Condition Based on k-Means Algorithm

作　　者：陆思宇 季盛[1] 文逸彦[1] LU Siyu;JI Sheng;WEN Yiyan(State Key Laboratory of Navigation and Safety Technology,Shanghai Ship and Shipping Research Institute Co.,Ltd.,Shanghai 200135,China)

机构地区：[1]上海船舶运输科学研究所有限公司航运技术与安全国家重点实验室,上海200135

出　　处：《上海船舶运输科学研究所学报》2023年第1期24-29,共6页Journal of Shanghai Ship and Shipping Research Institute

基　　金：中国远洋海运集团有限公司2019年科研专项(2019-1-H-001)。

摘　　要：为准确划分船舶主机运行工况,提升船舶主机性能监测能力,设计一种基于k-Means算法的船舶主机工况二次划分方法。基于实船运营大数据,分2个阶段开展主机运行工况划分研究,其中:第一阶段,以主机转速和功率为特征参数,初步划分主机运行工况;第二阶段,以海水温度和扫气箱温度为特征参数,对第一阶段划分的工况进行第二次划分。以某大型散货船为例,通过实船试验对该方法的有效性进行验证。试验结果表明,该方法能对复杂的船舶主机运行工况进行有效划分,可为船舶主机性能监测提供状态质量评估和辅助决策,有利于船舶主机健康管理和故障诊断工作的开展。To categorize the working condition that a main engine is in is a way to improve the performance monitoring ability.A two-stage working condition classification method for ship main engine based on k-Means algorithm is devised.In the first stage,the working conditions of the main engine are categorized according to the speed and power;In the second stage,each category of working conditions defined in the first stage is further divided into subcategories according to the seawater temperature and scavenging box temperature.The working condition of the main engine on a large bulk carrier is examined to verify the classification system.Experimental results show that the designed method can effectively identify the complex working conditions of ship main engine.This method can provide condition quality evaluation and decision-making assistance for ship main engine performance monitoring and is conducive to its health management and fault diagnosis.

关 键 词：智能船舶 船舶柴油机 工况划分 K-MEANS算法 主机性能 

分 类 号：U664.121[交通运输工程—船舶及航道工程]