航空发动机润滑系统故障知识图谱构建及应用

吴闯; 张亮; 唐希浪; 崔利杰; 谢小月

doi:10.13700/j.bh.1001-5965.2022.0434

航空发动机润滑系统故障知识图谱构建及应用

doi: 10.13700/j.bh.1001-5965.2022.0434

吴闯^{1, 2},
张亮¹,
唐希浪^1, ,,
崔利杰¹,
谢小月¹

1.
空军工程大学装备管理与无人机工程学院，西安 710051
2.
解放军95478部队，重庆 401329

基金项目: 国家自然科学基金(72201276)；中国博士后科学基金(2021M693941)；西安市科协青年人才托举计划项目(959202313098)

详细信息

通讯作者:
E-mail：tangxilang@sina.com

中图分类号: V239
计量
- 文章访问数: 255
- HTML全文浏览量: 121
- PDF下载量: 8
- 被引次数: 0
出版历程
- 收稿日期: 2022-05-28
- 录用日期: 2022-08-26
- 网络出版日期: 2022-09-22
- 整期出版日期: 2024-04-29

Construction and application of fault knowledge graph for aero-engine lubrication system

1.
College of Equipment Management and Unmanned Aerial Vehicle Engineering，Air Force Engineering University，Xi’an 710051，China
2.
95478 of the PLA，Chongqing 401329，China

Funds: National Natural Science Foundation of China (72201276); China Postdoctoral Science Foundation (2021M693941); Xi’an Association for Science and Technology Youth Talent Support Program (959202313098)

More Information

Corresponding author: E-mail：tangxilang@sina.com

摘要

摘要:
由于航空发动机润滑系统结构功能复杂，基于现有的健康管理系统开展故障诊断存在可解释性不足及高度依赖专家经验的问题，提出一套面向航空发动机润滑系统的故障知识图谱构建方法。在结合专家知识设计润滑系统故障知识图谱本体概念的基础上，采用双向长短期记忆(BiLSTM)神经网络和条件随机场(CRF)等深度学习技术实现知识自主抽取，并基于余弦距离和Jaccard相关系数法进行多源异构故障知识的融合。同时，基于构建的润滑系统故障知识图谱，实现润滑系统故障知识智能问答和故障归因分析应用。结果表明：知识图谱技术能够实现对润滑系统故障先验知识利用及故障原因解释，在智能故障诊断领域具有良好应用前景。
- 航空发动机 /
- 润滑系统 /
- 知识图谱 /
- 深度学习 /
- 知识问答
Abstract:
Due to the complex structure and function of aero-engine lubrication system, the existing health management system lacks sufficient interpretability and relies heavily on expert experience for fault diagnosis A method for constructing aero-engine lubrication system fault knowledge graph was proposed in this paper. By integrating expert knowledge, the concept of lubrication system fault knowledge graph ontology was designed. With the help of deep learning techniques such as bi-directional long short-term memory (BiLSTM) and conditional random field (CRF), we achieved the automatic extraction of unstructured knowledge. Next, based on the Cosine Distance and Jaccard coefficient, multi-source heterogeneous fault knowledge fusion was realized. In the end, incorporating the constructed areo-engine lubrication system fault knowledge graph, we achieved intelligent question and answer capabilities for lubrication system fault knowledge. The application results show that the knowledge graph technology can realize the utilization of prior knowledge of lubrication system faults and the explanation of fault causes, and has a good application prospect in the field of intelligent fault diagnosis.
- aero-engine /
- lubrication system /
- knowledge graph /
- deep learning /
- knowledge questions and answers

HTML全文

图 1 本文方法框架

Figure 1. Framework of proposed method

下载: 全尺寸图片幻灯片

图 2 模式层本体设计

Figure 2. Pattern layer ontology design

下载: 全尺寸图片幻灯片

图 3 润滑系统故障实体识别模型

Figure 3. Model of lubrication system fault entity recognition

下载: 全尺寸图片幻灯片

图 4 实体标注示例

Figure 4. Entity tagging example

下载: 全尺寸图片幻灯片

图 5 润滑系统故障关系抽取模型

Figure 5. Model of lubrication system fault relationship extraction

下载: 全尺寸图片幻灯片

图 6 故障根因挖掘

Figure 6. Root cause of fault

下载: 全尺寸图片幻灯片

图 7 故障文本标注可视化

Figure 7. Visualization of failure text annotation

下载: 全尺寸图片幻灯片

图 8 抽取模型整体评估

Figure 8. General comment of extraction model

下载: 全尺寸图片幻灯片

图 9 润滑系统故障知识图谱可视化

Figure 9. Visualization of lubrication system fault knowledge graph

下载: 全尺寸图片幻灯片

图 10 基于润滑系统故障知识图谱的知识问答

Figure 10. Knowledge Q&A based on lubrication system fault knowledge graph

下载: 全尺寸图片幻灯片

图 11 基于润滑系统故障知识图谱的故障归因分析

Figure 11. Analysis of fault-cause based on lubrication system fault knowledge graph

下载: 全尺寸图片幻灯片

表 1 实体相似度计算实例

Table 1. Examples of entity similarity computation

实体1	实体2	D_cosine	D_Jaccard
附件机匣	发附机匣	0.75	0.6
油液分析	油液化验	0.50	0.33
振动值超标	振动值异常	0.60	0.42
滑油泵组	滑油泵	0.87	0.75
滑油压力大	滑油压差大	0.80	0.66
金属屑信号器	金属末信号器	0.83	0.71

下载: 导出CSV

表 2 特征核心词

Table 2. Extraordinary key-word

问题分类	特征核心词
原因类	起因、导致、条件、故障树
事实类	组成、连接、是什么、系统、属性
统计类	有多少、集合、统计、总计
解决方案类	怎么办、处置、拆装、监控
预防措施类	提前、应急、保养、预先

下载: 导出CSV

表 3 润滑系统故障报告语料示例

Table 3. Corpus example of lubrication system failure report

情况	语料
简要情况	2022年X月X日，XX单位在执行飞行任务时，空中报发动机“滑油金属含量超标”告警，飞行任务停止，飞行员降低发动机转速后安全着陆。
排查情况	①着陆后，进行发动机滑油光谱分析，发现铁元素浓度值超标；②进一步孔探该发附件机匣内齿轮齿面异常磨损；③检查上次柔轴安装记录发现安装尺寸不够标准。
装备情况	发动机：型号XX；号码XXX；规定寿命XX；总使用时间XX小时XX分；制造厂XXX。附件机匣：型号XX；号码XXX；规定寿命XX；总使用时间XX小时XX分；制造厂XXX。
维修措施	①更换附件机匣；②故障的附件机匣返厂检查修理；③更换机匣后，清洗润滑系统，更换滑油3次。

下载: 导出CSV

表 4 润滑系统故障知识图谱实体数量

Table 4. Entity statistic numbers of lubrication system fault knowledge graph 个

故障模式	故障原因	排查方法	维修措施	预防措施	系统
295	331	240	149	93	3

故障案例	故障影响	故障时间	故障单位	故障单元
262	7	169	20	143

下载: 导出CSV

表 5 润滑系统故障知识图谱关系

Table 5. Relationship statistics of lubrication system fault knowledge graph 条

失效原因	发生时间	采取	引起	预防	造成
459	225	212	253	15	214

发生单位	组成	发现	关联	检查	发生
222	198	257	188	116	395

下载: 导出CSV

参考文献(26)

[1]	程荣辉, 张志舒, 陈仲光. 第四代战斗机动力技术特征和实现途径[J]. 航空学报, 2019, 40(3): 022698. CHENG R H, ZHANG Z S, CHEN Z G. Technical characteristics and implementation of the fourth-generation jet fighter engines[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(3): 022698 (in Chinese).
[2]	于皓, 张杰, 吴明辉, 等. 领域知识图谱快速构建和应用框架[J]. 智能系统学报, 2021, 16(5): 871-884. YU H, ZHANG J, WU M H, et al. A framework for rapid construction and application of domain knowledge graphs[J]. CAAI Transactions on Intelligent Systems, 2021, 16(5): 871-884 (in Chinese).
[3]	马忠贵, 倪润宇, 余开航. 知识图谱的最新进展、关键技术和挑战[J]. 工程科学学报, 2020, 42(10): 1254-1266. MA Z G, NI R Y, YU K H. Recent advances, key techniques and future challenges of knowledge graph[J]. Chinese Journal of Engineering, 2020, 42(10): 1254-1266 (in Chinese).
[4]	AMIT S. Introducing the knowledge graph: Things, not strings[EB/OL]. (2012-05-16)[2022-05-27]. http://www.blog.goole/products/search/introducing-knowledge-graph-things-not/.
[5]	侯梦薇, 卫荣, 陆亮, 等. 知识图谱研究综述及其在医疗领域的应用[J]. 计算机研究与发展, 2018, 55(12): 2587-2599. HOU M W, WEI R, LU L, et al. Research review of knowledge graph and its application in medical domain[J]. Journal of Computer Research and Development, 2018, 55(12): 2587-2599 (in Chinese).
[6]	王骏东, 杨军, 裴洋舟, 等. 基于知识图谱的配电网故障辅助决策研究[J]. 电网技术, 2021, 45(6): 2101-2112. WANG J D, YANG J, PEI Y Z, et al. Distribution network fault assistant decision-making based on knowledge graph[J]. Power System Technology, 2021, 45(6): 2101-2112 (in Chinese).
[7]	OU Q H, ZHENG W J, QI W W, et al. Research on the construction method of knowledge graph for electric power wireless private network[C]//Proceedings of the 2020 IEEE 10th International Conference on Electronics Information and Emergency Communication. Piscataway: IEEE Press, 2020: 10-13.
[8]	杜志强, 李钰, 张叶廷, 等. 自然灾害应急知识图谱构建方法研究[J]. 武汉大学学报(信息科学版), 2020, 45(9): 1344-1355. DU Z Q, LI Y, ZHANG Y T, et al. Knowledge graph construction method on natural disaster emergency[J]. Geomatics and Information Science of Wuhan University, 2020, 45(9): 1344-1355 (in Chinese).
[9]	陶坤旺, 赵阳阳, 朱鹏, 等. 面向一体化综合减灾的知识图谱构建方法[J]. 武汉大学学报(信息科学版), 2020, 45(8): 1296-1302. TAO K W, ZHAO Y Y, ZHU P, et al. Knowledge graph construction for integrated disaster reduction[J]. Geomatics and Information Science of Wuhan University, 2020, 45(8): 1296-1302 (in Chinese).
[10]	曹明, 王鹏, 左洪福, 等. 民用航空发动机故障诊断与健康管理: 现状、挑战与机遇-地面综合诊断、寿命管理和智能维护维修决策[J]. 航空学报, 2022, 43(9): 625574. CAO M, WANG P, ZUO H F, et al. Civil aero-engine diagnostics & health management: Current status, challenges and opportunities-comprehensive off-board diagnosis、life management & intelligent condition based MRO[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(9): 625574 (in Chinese).
[11]	韩涛, 黄海松, 姚立国. 面向航空发动机故障知识图谱构建的实体抽取[J]. 组合机床与自动化加工技术, 2021(10): 69-73. HAN T, HUANG H S, YAO L G. Entity extraction for aero-engine fault knowledge graph[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2021(10): 69-73 (in Chinese).
[12]	邱凌, 张安思, 李少波, 等. 航空制造知识图谱构建研究综述[J]. 计算机应用研究, 2022, 39(4): 968-977. QIU L, ZHANG A S, LI S B, et al. Advances in building knowledge graphs for aerospace manufacturing[J]. Application Research of Computers, 2022, 39(4): 968-977 (in Chinese).
[13]	聂同攀, 曾继炎, 程玉杰, 等. 面向飞机电源系统故障诊断的知识图谱构建技术及应用[J]. 航空学报, 2022, 43(8): 625499. NIE T P, ZENG J Y, CHENG Y J, et al. Knowledge graph construction technology and its application in aircraft power system fault diagnosis[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(8): 625499 (in Chinese).
[14]	王昊奋, 漆桂林, 陈华钧. 知识图谱方法、实践与应用[M]. 北京: 电子工业出版社, 2019: 425-427. WANG H F, QI G L, CHEN H J. Knowledge graph method, practice and application[M]. Beijing: Publishing House of Electronics Industry Press, 2019: 425-427 (in Chinese).
[15]	陈永当, 杨海成, 杜兵劳. 基于本体的航空发动机设计知识组织模型构建与分析[J]. 航空动力学报, 2007, 22(1): 90-95. CHEN Y D, YANG H C, DU B L. Building and analysis of ontology-based knowledge organization model for aircraft engine design[J]. Journal of Aerospace Power, 2007, 22(1): 90-95 (in Chinese).
[16]	朱庆, 王所智, 丁雨淋, 等. 铁路隧道钻爆法施工智能管理的安全质量进度知识图谱构建方法[J]. 武汉大学学报(信息科学版), 2022, 47(8): 1155-1164. ZHU Q, WANG S Z, DING Y L, et al. Construction method of "safety-quality-schedule" knowledge graph for intelligent management of drilling and blasting construction of railway tunnels[J]. Geomatics and Information Science of Wuhan University, 2022, 47(8): 1155-1164 (in Chinese).
[17]	李震, 刘斌, 苗虹, 等. 基于本体的软件安全性需求建模和验证[J]. 北京航空航天大学学报, 2012, 38(11): 1445-1449. LI Z, LIU B, MIAO H, et al. Modeling and verification of software safety requirement based on ontology[J]. Journal of Beijing University of Aeronautics and Astronautics, 2012, 38(11): 1445-1449 (in Chinese).
[18]	刘峤, 李杨, 段宏, 等. 知识图谱构建技术综述[J]. 计算机研究与发展, 2016, 53(3): 582-600. LIU Q, LI Y, DUAN H, et al. Knowledge graph construction techniques[J]. Journal of Computer Research and Development, 2016, 53(3): 582-600 (in Chinese).
[19]	XU K, ZHOU Z F, HAO T Y, et al. A bidirectional LSTM and conditional random fields approach to medical named entity recognition [C]//Proceedings of the International Conference on Advanced Intelligent Systems and Informatics. Berlin: Springer, 2017: 355-365.
[20]	DEVLIN J, CHANG M W, LEE K, et al. Bert: pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Minneapolis: Association for Computational Linguistics, 2019: 4171-4186.
[21]	吴赛赛, 周爱莲, 谢能付, 等. 基于深度学习的作物病虫害可视化知识图谱构建[J]. 农业工程学报, 2020, 36(24): 177-185. WU S S, ZHOU A L, XIE N F, et al. Construction of visualization domain-specific knowledge graph of crop diseases and pests based on deep learning[J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(24): 177-185 (in Chinese).
[22]	罗欣, 陈艳阳, 耿昊天, 等. 基于深度强化学习的文本实体关系抽取方法[J]. 电子科技大学学报, 2022, 51(1): 91-99. LUO X, CHEN Y Y, GENG H T, et al. Entity relationship extraction from text data based on deep reinforcement learning[J]. Journal of University of Electronic Science and Technology of China, 2022, 51(1): 91-99 (in Chinese).
[23]	赵晓娟, 贾焰, 李爱平, 等. 多源知识融合技术研究综述[J]. 云南大学学报(自然科学版), 2020, 42(3): 459-473. ZHAO X J, JIA Y, LI A P, et al. A survey of the research on multi-source knowledge fusion technology[J]. Journal of Yunnan University (Natural Sciences Edition), 2020, 42(3): 459-473 (in Chinese).
[24]	乔骥, 王新迎, 闵睿, 等. 面向电网调度故障处理的知识图谱框架与关键技术初探[J]. 中国电机工程学报, 2020, 40(18): 5837-5849. QIAO J, WANG X Y, MIN R, et al. Framework and key technologies of knowledge-graph-based fault handling system in power grid[J]. Proceedings of the CSEE, 2020, 40(18): 5837-5849 (in Chinese).
[25]	董丽叶. 基于知识图谱的S市地铁机电设备故障处理优化研究[D]. 石家庄: 河北科技大学, 2020: 43-48. DONG L Y. Fault treatment optimization of subway electromechanical equipment in S city based on knowledge graph[D]. Shijiazhuang: Hebei University of Science and Technology, 2020: 43-48 (in Chinese).
[26]	ARTSTEIN R, POESIO M. Inter-coder agreement for computational linguistics[J]. Computational Linguistics, 2008, 34(4): 555-596 . doi: 10.1162/coli.07-034-R2