针对舰船离心泵类系统建造阶段缺乏声学故障诊断方法的问题，本文在振动传递路径分析及Noisy-Max假设的基础上提出一种基于贝叶斯网络的舰船离心泵类系统声学故障诊断方法。通过陆上台架模拟试验确定了贝叶斯网络模拟中的待定参数，并通过随机选取的5组试验数据对该方法的有效性进行验证。结果表明，该方法诊断结果与实际故障吻合，诊断准确率较高。
Aiming at the lack of acoustic fault diagnosis method in the construction stage of ship centrifugal pump system, an acoustic fault diagnosis method of ship centrifugal pump system based on Bayesian network is proposed on the basis of vibration transmission path analysis and Noisy-Max hypothesis. The undetermined parameters in Bayesian network simulation are determined through the shelf simulation test, and the effectiveness of the method is verified by five groups of randomly selected test data. The results show that the diagnosis results of this method are consistent with the actual faults, and this method has high diagnosis accuracy.
2022,44(10): 93-96 收稿日期：2022-01-07
DOI：10.3404/j.issn.1672-7649.2022.10.018
分类号：U674.7
作者简介：李春光(1987-)，男，硕士，工程师，研究方向为舰船总体
参考文献：
[1] O'MARA D, GEAR G, MCMAHON D. Automated self-noise monitoring and acoustic fault detection[C]// UDT Korea, 2002.
[2] 章林柯, 何琳, 江涌. 基于MIMO模型的潜艇声学故障预报研究[J]. 武汉理工大学学报, 2008, 32(2): 320–322+326
[3] PEARL J. Bayesian networks: A model of self-activated memory for evidential reasoning[J]. Proceeding of Cognitive Science Society, 1985: 1985
[4] CAI Z , SUN S , SI S , et al. Research of failure prediction Bayesian network model[C]// International Conference on Industrial Engineering & Engineering Management. IEEE, 2009.
[5] 李前舸. 基于诊断贝叶斯网络的螺杆式冷水机组故障检测与诊断研究[D]. 上海: 上海交通大学, 2020.
[6] 邹树梁, 陈树强. 基于故障树分析和贝叶斯网络方法的液压举升机构故障诊断[J]. 南华大学学报:自然科学版, 2021, 35(4): 6
[7] 张皓宇, 付锦洲, 魏曙寰. 基于贝叶斯网络的舰艇机电装备故障分析方法研究[J]. 装备制造技术, 2020(12): 5
[8] 张义忠, 张华栋, 计方, 等. 船舶典型通海电动泵管路系统建造声学故障模式初探[J], 舰船科学技术, 2021, 43(S1): 7–12.
[9] 陈东超. 基于贝叶斯网络的汽轮发电机组故障诊断方法及应用研究[D]. 北京: 华北电力大学, 2018.
[10] 张连文, 郭海鹏. 贝叶斯网引论[M]. 北京: 科学出版社, 2006: 41–45.
