基于误差预测修正的液体火箭发动机故障预测方法研究

推进技术 ›› 2016, Vol. 37 ›› Issue (8): 1569-1578.

基于误差预测修正的液体火箭发动机故障预测方法研究

聂侥,吴建军

国防科学技术大学航天科学与工程学院，湖南长沙 410073,国防科学技术大学航天科学与工程学院，湖南长沙 410073

发布日期:2021-08-15
作者简介:聂侥，男，博士生，研究领域为液体火箭发动机健康监控。
基金资助:
国家自然科学基金(51206181；51506219)。

Research of Liquid-Propellant Rocket Engine Failure Prediction Method Based on Error Predicting Correlation

College of Aerospace Science and Engineering，National University of Defense Technology，Changsha 410073，China and College of Aerospace Science and Engineering，National University of Defense Technology，Changsha 410073，China

Published:2021-08-15

摘要/Abstract

摘要： 为解决液体火箭发动机故障预测这一难题，提出一种基于误差预测修正的故障预测方法。在历史数据的基础上建立小波过程神经网络故障预测模型，同步计算学习样本的预测误差，根据上述误差建立双并联离散过程神经网络预测模型。预测时，将预测误差值实时补偿到小波过程神经网络预测模型以提高预测精度。通过液体火箭发动机地面试验中的涡轮泵数据对该方法进了验证。结果表明，该方法在预测精度和适应能力上较单一的过程神经网络预测模型有显著提高，进行10步预测时，预测值的标准化均方根误差为0.392，预测平均耗时为76ms，能够用于解决液体火箭发动机故障预测问题。

关键词: 液体火箭发动机；故障预测；小波过程神经网络；双并联离散过程神经网络；误差修正

Abstract: In order to solve the problem of liquid-propellant rocket engine (LRE) failure prediction accurately，reliably and rapidly，a failure prediction method based on error predicting correction is proposed. The prediction model of wavelet process neural network (WPNN)，established based on historical data，is firstly adopted to make failure prediction of liquid-propellant rocket engine. The prediction error of every sample is recorded for setting up error prediction model，which is a double parallel feedforward discrete input process neural network (DPFDPNN). Then the prediction error is estimated with DPFDPNN model to compensate the predicted value with the WPNN model. Finally，the proposed method is validated through a test case with ground testing data of LRE hydrogen turbopump. The results show that，the proposed method outperformed the WPNN model at accuracy and adaptability. The normalized root mean square error (NRMSE) of the prediction value is 0.392，when the number of prediction step is 10，and the average computing time is 76ms. The proposed method may be used to solve the problem of LRE failure prediction.

Key words: Liquid-propellant rocket engine；Failure prediction；Wavelet process neural networks；Double parallel feedforward discrete input process neural network；Error correlation

聂侥,吴建军. 基于误差预测修正的液体火箭发动机故障预测方法研究[J]. 推进技术, 2016, 37(8): 1569-1578.

NIE Yao,WU Jian-jun. Research of Liquid-Propellant Rocket Engine Failure Prediction Method Based on Error Predicting Correlation[J]. Journal of Propulsion Technology, 2016, 37(8): 1569-1578.

参考文献

[1] Gao H, Keyou Y. Trends in Extreme Learning Machines: A Review[J]. Neural Networks, 2015, 61(2): 32-48.
[2] 彭宇, 刘大同, 彭喜元. 故障预测与健康管理技术综述[J]. 电子测量与仪器学报, 2010, 24(1): 1-9.
[3] Pavel L, Christian G. Incremental Support Vector Learning: Analysis, Imple Mentation and Applications[J]. Journal of Machine Learning Research, 2006, 7:1909-1936.
[4] 田路, 张炜, 杨正伟. Elman型神经网络在液体火箭发动机故障预测中的应用[J]. 弹箭与制导学报, 2009, 29(1):191-194.
[5] 陈世立, 陈新民. 改进 BP神经网络在冲压发动机性能预测中的应用[J]. 导弹与航天运载技术, 2007, 3(298): 45-49.
[6] 王晔, 段志信. 基于 Matlab和 BP神经网络的固体火箭发动机比冲性能的预测[J]. 内蒙古科技与经济, 2007, 8(138):73-74.
[7] 田干, 张炜, 杨正伟, 等. SVM方法在火箭发动机故障预测中的应用研究[J]. 机械科学与技术, 2010, 29(1): 63-67.
[8] XU Shao-hua, HE Xin-gui. Some Theoretical Issues on Continuous Process Neural Networks[J]. Acta Electronica Sinic, 2006, 34(10): 838-1841.
[9] HUANG Guang-bin, DING Xiao-jian. Optimization Method Based on Extreme Learning Machine for Classification[J]. Neurocomputing, 2010, 74(1):155-163.
[10] DING Gang, ZHONG Shi-sheng. Aircraft Engine Lubricating Oil Monitoring by Process Neural Networks[J]. Neural Network World, 2006, 16(1): 15-24.
[11] DING Gang, ZHONG Shi-sheng. Convergence Analysis of the Learning Algorithm for Parallel Feed-Forward Process Neural Networks[J]. Journal of Computational Information Systems, 2005, 1(3): 477-483.
[12] YU Guang-bin, DING Gang, LIN Lin. Aircraft Engine Fuel Flow Prediction Using Process Neural Network[J]. International Journal of Control and Automation, 2015, 7(3): 53-61.
[13] 钟诗胜, 李洋. 基于小波过程神经网络的飞机发动机状态监视[J]. 航空学报, 2007, 28(1): 68-71.
[14] 金向阳, 林琳, 钟诗胜. 航空发动机振动趋势预测的过程神经网络法[J]. 振动、测试与诊断, 2011, 31(331-336).
[15] 刘菲菲, 彭荻, 贺彦林, 等. 基于极限学习的过程神经网络研究及化工应用[J]. 上海交通大学学报, 2014, 48(7): 977-981.
[16] LIU Z G, Wang X R, Qian Q Q. A Review of Wavelet Net-Vorks and Their Applications[J]. Automation of Electric Power System. 2003, 27(6): 73-79.
[17] DING Gang, Zhong Shi-Sheng, LI Yang. Time Series Prediction Using Wavelet Process Neural Network[J]. Chinese Phys.B, 2008, 17(6): 1674-1056.
[18] Hagan M T, Menhaj M B. Training Feedforward Networks with the Marquardt Algorithm [J]. IEEE Transactions on Neutral Networks, 1994, 5(6): 989-993.
[19] 钟诗胜, 丁刚, 付旭云. 过程神经网络模型及其工程应用[M] 北京:国防工业出版社, 2014.
[20] LIANG Nan-ying, HUANG Guang-bin, Senior M. A Fast and Accurate Online Sequential Learning Algorithm for Feedforward Networks[J]. IEEE Transactions on Neutral Networks, 2006, 17(6): 1411-1423.
[21] Zhang R, Lan Y, Huang G B. Universal Approximation of Extreme Learning Machine with Adaptive Growth of Hidden Nodes[J]. IEEE Transactions on Neutral Networks, 2012, 23(2): 365-371.
[22] 谢廷峰. 液体火箭发动机健康监控关键技术研究[D]. 长沙:国防科学技术大学, 2008.
[23] 夏鲁瑞. 液体火箭发动机涡轮泵健康监控关键技术及系统研究[D]. 长沙:国防科学技术大学, 2010.
[24] 雷达, 钟诗胜. 用集成过程神经网络预测民航发动机振动趋势[J]. 振动、测试与诊断, 2015, 35(1):137-141.
[25] 夏鲁瑞, 胡茑庆, 秦国军. 转速波动状态下涡轮泵典型故障诊断方法[J]. 推进技术, 2009, 30(3): 342-346. (XIE Lu-rui, HU Niao-qing, QING Guo-jun. Diagnosis Method for Turbopump Typical Fault Under Condition of Speed Fluctuation[J]. Journal of Propulsion Technology, 2009, 30(3): 342-346.)
[26] 谢光军, 胡茑庆. 涡轮泵故障检测系统[J]. 推进技术. 2006, 27(2): 141-145. (XIE Guang-jun, HU Niao-qing. Turbopump Fault Detection System[J]. Journal of Propulsion Technology, 2006, 27(2): 141-145.)
[27] 张炜, 张玉祥, 黄先祥. 基于神经网络的涡轮泵多故障诊断[J]. 推进技术, 2003, 24(1): 17-20. (ZHANG Wei, ZHANG Yu-xiang, HUANG Xian-xiang. Multi-Fault Diagnosis for Turbo-Pump Based on Neural Network[J]. Journal of Propulsion Technology, 2003, 24(1): 17-20.)
[28] Jacek M leski. On Support Vector Regression Machines with Linguistic Interpretation of the Kernel Matrix[J]. Fuzzy Sets and Systems, 2006, 157(8): 1092-1113. 　　　　　　　　　　　　　*　收稿日期:2015-10-03；修订日期:2016-01-27。基金项目:国家自然科学基金(51206181；51506219)。作者简介:聂侥,男,博士生,研究领域为液体火箭发动机健康监控。E-mail: nieyao121@163.com(编辑:朱立影)