船舶动力装置状态参数动态阈值方法

姜兴家; 张鹏; 孙晓磊; 孙培廷; 张跃文

doi:10.16411/j.cnki.issn1006-7736.2018.04.005

大连海事大学学报 >

2018 , Vol. 44 >Issue 4: 28 - 34

DOI: https://doi.org/10.16411/j.cnki.issn1006-7736.2018.04.005

船舶与海洋工程

船舶动力装置状态参数动态阈值方法

姜兴家 ,
张鹏 ,
孙晓磊 ,
孙培廷 ,
张跃文

展开

（大连海事大学轮机工程学院，辽宁大连 116026）

姜兴家（1982—），男，硕士，讲师，E-mail：jiangxj@dlmu.edu.cn.

收稿日期: 2018-04-10

修回日期: 2018-05-10

网络出版日期: 2018-05-10

基金资助

中央高校基本科研业务费专项资金资助项目（3132016357）；中国船级社智能诊断项目（9215294）.

收起

Dynamic threshold method for state parameters of ship power plant

JIANG Xing-jia ,
ZHANG Peng ,
SUN Xiao-lei ,
SUN Pei-ting ,
ZHANG Yue-wen

Expand

(Marine Engineering College, Dalian Maritime University, Dalian 116026,China)

Received date: 2018-04-10

Revised date: 2018-05-10

Online published: 2018-05-10

Supported by

Fold

摘要

针对单一阈值判断法在判断船舶动力装置状态参数时的局限性问题，提出阈值走廊的概念.利用麦夸特算法对阈值基线进行非线性拟合，运用标准化残差中的质量控制原则设定阈值走廊，使用平均绝对百分比误差法对阈值基线进行比较并建立阈值自适应模型，使得船舶动力装置在性能退化过程中能够实现阈值走廊的自更新.选取“育鲲”轮某航期内海水泵出口压力进行实例验证.结果表明，压力阈值随海水滤器性能退化而发生变化；建立的阈值走廊能够显示阈值变化趋势并能灵活精确地反应不同时段的阈值范围.

关键词： 船舶动力装置; 状态参数; 阈值走廊; 麦夸特; 标准化残差; 阈值自适应

本文引用格式

姜兴家 , 张鹏 , 孙晓磊 , 孙培廷 , 张跃文 . 船舶动力装置状态参数动态阈值方法[J]. 大连海事大学学报, 2018 , 44(4) : 28 -34 . DOI: 10.16411/j.cnki.issn1006-7736.2018.04.005

Abstract

In order to improve the limitation of the single threshold judgement method in determining the state parameters of the ship’s power plant, the concept of threshold corridor was proposed. The threshold baseline was nonlinear fitted by using the Marquardt algorithm, and the threshold corridor was set with the quality control principle of standardized residuals. The mean absolute percentage error (MAPE) method was used to compare the threshold baselines and establish the threshold adaptive model to realize the selfrenewal of the threshold corridor during the performance degradation process of the power plant. Selecting the sea water pump outlet pressure of MV “YUKUN” for verification, and result shows that the pressure threshold changes with the degradation of seawater filter, and the threshold corridor can display the trend of the threshold and reflect the range of the threshold flexibly and accurately in different time periods.

Key words： marine power plant; state parameters; threshold corridor; Marquardt; standardized residual; threshold adaptive

参考文献

[1]李毅中.中国版工业4.0来了:期待中国制造大变身[J].工程机械, 2015. [2]高宗江, 张英俊, 孙培廷, 等.无人驾驶船舶研究综述[J].大连海事大学学报, 2017, 43(2):1-7 [3]季策, 郭伟, 张超.基于动态功率阈值降低的部分传输序列算法[J].系统工程与电子技术, 2017, 39(7):1620-1624 [4]张阿漫, 周其新, 姚熊亮, 等.基于船体与设备一体化的设备抗冲击阈值[J].爆炸与冲击, 2009, 29(4):375-379 [5]程晶, 那建, 杨会金.舰船设备振动监测阈值设定方法[J].舰船科学技术, 2012, 34(11):68-70 [6]陈子建, 潘璠, 王振涛.基于变阈值的船舶动力装置状态监测与评估研究[J].Mechanical Engineering\s& \stechnology, 2014, 03(4):144-151 [7]Lu J, Ding D, Wu L, et al.Method of Fire Image Identification Based on Optimization Theory[J].系统工程与电子技术英文版, 2002, 13(2):78-83 [8].American Society of Mechanical Engineers，ASME Boiler and Pressure Vessel Code，Section Ⅺ，Ap-pendix G，“Rules for Inservice Inspection of Nuclear Power Plant Components”［S］． 2007 Edition，ASME，New York． [9]Ren L, Yong G.Applicability of the Revised Mean Absolute Percentage Errors (MAPE) Approach to Some Popular Normal and Non-normal Independent Time Series[J].International Advances in Economic Research, 2009, 15(4):409-420 [10]Baczkowski B M, Johnstone T, Walter H, et al.Sliding-window analysis tracks fluctuations in amygdala functional connectivity associated with physiological arousal and vigilance during fear conditioning.[J]. Neuroimage, 2017, 153.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献