| 43 | 0 | 5 |
| 下载次数 | 被引频次 | 阅读次数 |
针对飞行器惯导系统测试影响、数据模态不均衡和专家知识不确定等对惯导系统健康评估精度产生影响的问题，构建了一种考虑测试履历影响的置信规则库模型（Belief Rule Base with Test Impact,BRB-T）。首先，通过引入累计测试次数与通电时长来反映测试履历对惯导系统健康评估的影响，提出一种基于自度量的测试履历影响转换方法，从而将测试次数和通电时长2个不同量级指标转换到统一度量框架下。然后，基于所构建的BRB-T模型，提出了一种考虑测试履历影响的飞行器惯导系统健康状态评估方法，以提高飞行器惯导系统在多次通电测试下的健康评估精度。其中，在所构建的BRB-T模型中，通过专家规则构建过程实现专家知识的嵌入，并通过少量非均衡数据对模型参数进行优化调整，实现了知识数据互补增强。最后，利用某型号飞行器惯导系统对所提出方法的有效性开展了实验验证。结果表明：该方法的健康评估误差为0.0131，相比于神经网络、模糊推理等数据/知识驱动方法的精度分别提高了59.06%和69.61%，证明了所提方法的有效性。
Abstract：To address the influences of test impact, data model imbalance, and expert knowledge uncertainty on the health assessment accuracy of INS(Inertial Navigation System), a BRBT(Belief Rule Base with Test Impact)model was proposed. First, the cumulative number of tests and power-on time were introduced to reflect the test impact on INS health assessments. A test impact conversion method based on self-measurement was proposed to convert the test frequency and power-on duration, which were indicators at two different levels, into a unified measurement framework. Subsequently, based on the BRB-T model, a new health assessment method for aircraft INS with consideration of test impact was proposed to improve the accuracy of the health assessment of aircraft INS under multiple power-on tests. In the BRB-T model, expert knowledge was achieved through the process of constructing expert rules. Complementary enhancement of knowledge data was achieved by optimizing and adjusting the model parameters using a small amount of disequilibrium data. Finally, the effectiveness of the proposed method was verified using the INS of a certain type of aircraft. The results show that the health assessment error of the proposed method is 0.0131, which improves the accuracy by 59.06% and 69.61% respectively compared with data/knowledge-driven methods, such as neural networks and fuzzy reasoning,demonstrating the effectiveness of the proposed method.
[1] FENG Z C, ZHOU Z C, YANG R H, et al.Fault-tolerant control based on belief rule base expert system for multiple sensors concurrent failure in liquid launch vehicle[J]. Nonlinear Dynamics,2023,111(5):4357-4373.
[2]赵宇，周志杰，樊红东，等．基于置信规则库的惯性导航系统性能评估方法[J]．战术导弹技术，2023(5):89-96.ZHAO Y,ZHOU Z J,FAN H D,et al.Performance evaluation method of inertial navigation system based on belief rule base[J]. Tactical Missile Technology,2023(5):89-96.
[3]赵丽华，程晓玉，高璐，等．液体火箭箭体结构的半定量安全状态评估方法[J]．哈尔滨理工大学学报，2023,28(5):61-67.ZHAO L H, CHENG X Y, GAO L, et al.Semi-quantitative safety assessment method for liquid rocket body structure[J]. Journal of Harbin University of Science and Technology, 2023, 28(5):61-67.
[4]江秀红．多态复杂系统的可靠性分析及维修策略研究[D]．大连：大连理工大学，2016:56-77.JIANG X H. Reliability analysis and maintenance strategy research of multi-state complex system[D]. Dalian:Dalian University of Technology,2016:56-77.
[5]曲芮娇．数据驱动的惯导系统剩余寿命预测方法研究[D]．北京：北京交通大学，2022:89-123.QU R J.Research on data-driven residual life prediction method of inertial navigation system[D].Beijing:Beijing Jiaotong University,2022:89-123.
[6]吴克雄，王振华．基于专家系统和数据驱动的健康评估分析方法[J]．黑龙江科学，2019,10(16):64-65.WU K X, WANG Z H. Health assessment and analysis method based on expert system and data driven[J]. Heilongjiang Science, 2019, 10(16):64-65.
[7]董昕昊，周志杰，胡昌华，等．基于分层置信规则库的惯导系统性能评估方法[J]．航空学报，2021,42(7):441-451.DONG X H,ZHOU Z J,HU C H,et al.Performance evaluation method for inertial system based on hierarchical belief rule base[J].Acta Aeronautica et Astronautica Sinica,2021,42(7):441-451.
[8] FENG Z C,ZHOU Z J,BAN X J,et al.Sensor fault diagnosis and tolerant control based on belief rule base for complex system[J]. Journal of Systems Science&Complexity, 2023, 36(3):1002-1023.
[9] ZHANG Q X, ZHAO B Y, HE W, et al. A behavior prediction method for complex system based on belief rule base with structural adaptive[J].Applied Soft Computing,2024,151:111118.
[10]杨隆浩，陈江鸿，叶菲菲，等．基于规则聚类和参数学习的扩展置信规则库推理模型[J]．控制与决策，2024,39(8):2685-2693.YANG L H, CHEN J H, YE F F, et al. Extended belief rule base inference model based on rule clustering andparameter learning[J]. Control and Decision,2024,39(8):2685-2693.
[11]常雷雷，徐晓滨，徐晓健．基于主导从属框架的变结构置信规则库多目标优化方法[J]．系统工程理论与实践，2022,42(2):514-526.CHANG L L, XU X B, XU X J. Multi-objective optimization method of variable-structured belief rule base using dominant subordinate framework[J].Systems Engineering Theory&Practice,2022,42(2):514-526.
[12] ZHANG C C,ZHOU Z,NING P,et al.MBRB:micro-belief rule Base model based on cautious conjunctive rule for interpretable fault diagnosis[J].Engineering Applications of Artificial Intelligence,2024,135:108598.
[13]HE W,CHENG X Y,ZHAO X.An interval construction belief rule base with interpretability for complex systems[J].Expert Systems with Applications,2023,229:120485.
[14]蒲俊，马清亮，顾凡，等．基于遗传算法的多项式不确定非线性系统保性能控制[J]．火箭军工程大学学报（自然科学版），2019:33(1):69-76.PU J,MA Q L,GU F,et al.GA-based design of guaranteed cost controller for polynomial uncertain nonlinear system[J]. Journal of Rocket Force University of Engineering(Natural Science Edition),2019,33(1):69-76.
[15]郑建飞，余铜辉，张会会，等．一种不可修复设备的备件订购和更换决策方法[J]．火箭军工程大学学报（自然科学版），2019,33(2):33-38.ZHENG J F,YU T H,ZHANG H H,et al.A decision-making method for spare parts ordering and replacement of non-repairable equipment[J].Journal of Rocket Force University of Engineering(Natural Science Edition),2019:33(2):33-38.
[16] ZHAO Y, ZHOU Z J, FAN H D, et al. A health state prediction model based on belief rule base and LSTM for complex systems[J]. Intelligent Automation&Soft Computing,2024,39(1):73-91.
[17] LIAN Z,ZHOU Z J,HU C H,et al.Interpretable large-scale belief rule base for complex industrial systems modeling with expert knowledge and limited data[J]. Advanced Engineering Informatics,2024,62:102852.
[18] LI S Z, LIU B X, FENG J Y, et al. Optimal maintenance decision method for a sensor network based on belief rule base considering attribute correlation[J]. International Journal of Intelligent Systems,2024,2024:6616366.
[19] WANG Z Y, HE W, YANG R H, et al. A new belief rule base based distributed online fault diagnosis method for multi-agent systems[J].Asian Journal of Control,2024,26(2):668-682.
[20]QIAN G Y, LI J Y, HE W, et al. An online intrusion detection method for industrial control systems based on extended belief rule base[J]. International Journal of Information Security,2024,23:2491-2514.
[21] WANG H B,GUAN X,YI X,et al.Heterogeneous information fusion recognition method based on belief rule structure[J].Journal of Systems Engineering and Electronics,2024,35(4):955-964.
[22] ZHANG Z J,DENG Q,HE W.A new method based on belief rule base with balanced accuracy and interpretability for student achievement prediction[J].Mathematics,2024,12(20):3283.
[23] LIU M,HE W,MA N,et al.A new reliability health status assessment model for complex systems based on belief rule base[J].Reliability Engineering&System Safety,2025,254:110614.
[24] QIAN H, PAN Y T, WANG X H, et al. Research on the optimization of belief rule bases using the Naive Bayes theory[J]. Frontiers in Energy Research,2024,12:1396841.
[25] ZHANG Q X,LI K L,ZHANG G L,et al.A complex system health state assessment method with reference value optimization for interpretable BRB[J].Scientific Reports,2024,14(1):2334.
基本信息:
DOI：10.20189/j.cnki.CN/61-1527/E.202502003
中图分类号:V249.3;TN96
引用信息:
[1]熊子龙,曹哲.考虑测试履历影响的惯导系统健康状态评估方法[J].火箭军工程大学学报,2025,39(02):22-28.DOI:10.20189/j.cnki.CN/61-1527/E.202502003.
基金信息:
国家自然科学基金（62203461,62203365）