Multi-objective performance optimization of turbofan engine for test run

面向检验试车的涡扇发动机多目标性能优化

¹ AECC Aviation Power Co., Ltd, Xi’an 710021, China
² School of Mechanical Engineering, Northwestern Polytechnical University, Xi’an 710072, China

Received: 2 September 2023

Abstract

Turbofan engines are widely used in military and civilian aviation fields due to their high propulsion efficiency and low fuel consumption rate, and their performance directly affects the safety and stability of flight mission. It is of great practical significance to optimize the turbine inlet temperature and high-pressure compressor speed under different thrust states, so as to improve the pass rate of the first test run. This paper proposes a multi-objective performance optimization framework for turbofan engines. On the historical production dataset of a certain type of turbofan engine, the turbine inlet temperature and high-pressure compressor speed under different thrusts are taken as target variables, and area variable a, area variable b, and angle variable c in the assembly stage are taken as attribute variables. Then, the multi-objective performance optimization model based on tree augmented naive bayes is established and compared and verified with the current mainstream algorithm for verification. Finally, combining with the posterior qualified probability inference and state combination global search method, a recommended state combination table is given to assist enterprises in the formulation of component production and manufacturing assembly standards, thereby optimizing turbofan engine performance, reducing reassembly requirements, and improving the pass rate of the first test run.

摘要

涡扇发动机因其高推进效率、低燃油消耗率等特点广泛应用于军民用飞机, 其性能直接影响飞行任务的安全与稳定。针对涡扇发动机不同状态下涡轮前温度与高压转速比指标进行性能优化, 从而提高其一次检验试车通过率, 具有重要现实意义。提出涡扇发动机的多目标性能优化框架, 在某型号涡扇发动机历史生产数据集上, 以检验试车过程中不同状态下涡轮前温度与高压转速比为目标变量, 以某面积a、某面积b、某角度c为属性变量, 建立涡扇发动机多目标性能模型并与目前主流算法模型对比验证, 最后结合通过检验试车的后验概率推理与状态组合全局搜索, 给出推荐状态组合表, 辅助企业制定零部件生产制造装配标准, 从而优化涡扇发动机性能、减少重新装配次数并提高一次检验试车通过率。

1. JIAO Huabin, MO Song. Present status and development trend of aircraft turbine engine[J]. Aeronautical Manufacturing Technology, 2015, 58(12): 62–65 (in Chinese) [Google Scholar]
2. GARG S. NASA glenn research in controls and diagnostics for intelligent aerospace propulsion systems[R]. NASA TM-2005-214036, 2005 [Google Scholar]
3. SUN Jianguo. Modern aviation power plant control[M]. Beijing: Aviation Industry Press, 2009 (in Chinese) [Google Scholar]
4. CHEN Guo. Forecasting engine performance trend by using structure self-adaptive neural network[J]. Acta Aeronautica et Astronautica Sinica, 2007, 28(3): 535–539 [Article] (in Chinese) [Google Scholar]
5. KIAKOJOORI S, KHORASANI K. Dynamic neural networks for gas turbine engine degradation prediction, health monitoring and prognosis[J]. Neural Computing and Applications, 2016, 27(8): 2157–2192 [Article] [CrossRef] [Google Scholar]
6. LI Jingfeng, CHEN Yunxiang, XIANG Huachun, et al.. Remaining useful life prediction for aircraft engine based on LSTM-DBN[J]. Systems Engineering and Electronics, 2020, 42(7): 1637–1644 (in Chinese) [Google Scholar]
7. ZHANG B, LI Y, BAI Y, et al.. Aeroengines remaining useful life prediction based on improved C-loss ELM[J]. IEEE Access, 2020, 8:49752–49764 [Article] [NASA ADS] [CrossRef] [Google Scholar]
8. ZHOU Tao, WANG Yongchao, ZHANG Xujing, et al.. Data-model interactive remaining useful life prediction of stochastic degrading devices based on deep feature fusion network[J]. Computer Integrated Manufacturing Systems, 2022, 28(12): 3937–3945 (in Chinese) [Google Scholar]
9. LIN Zhifu, XIAO Hong, WANG Zhanxue, et al.. An aeroengine digital engineering model based on physics-embedded neural networks[J]. Journal of Propulsion Technology, 2023, 44(11): 54–65 (in Chinese) [Google Scholar]
10. CHEN Yingwu, GAO Yanfang. Survey of extended Bayesian networks[J]. Control and Decision, 2008, 23(10): 1081–1086 [Article] (in Chinese) [Google Scholar]
11. CAI Z, GUO P, SI S, et al.. Analysis of prognostic factors for survival after surgery for gallbladder cancer based on a Bayesian network[J]. Scientific Reports, 2017, 7(1): 293 [Article] [CrossRef] [Google Scholar]
12. PETOUSIS P, HAN S X, ABERLE D, et al.. Prediction of lung cancer incidence on the low-dose computed tomography arm of the national lung screening trial: a dynamic Bayesian network, Artificial Intelligence in Medicine, 2016, 72: 42–55 [Article] [CrossRef] [Google Scholar]
13. ZHENG Wei, SHEN Wen, ZHANG Yingpeng. Implement spam filter by improving naïve Bayesian algorithm[J]. Journal of Northwestern Polytechnical University, 2010, 28(4): 622–627 [Article] (in Chinese) [Google Scholar]
14. LI Naixin, LU Zhong, ZHOU Jia. Reliability assessment based on Bayesian networks for electro hydraulic servo actuator[J]. Journal of Northwestern Polytechnical University, 2016, 34(5): 915–920 [Article] (in Chinese) [Google Scholar]
15. WANG Ning, WANG Yuhang, CAI Zhiqianget al.. Performance optimization scheme of turboshaft aeroengine based on Bayesian network[J]. Journal of Northwestern Polytechnical University, 2021, 39(2): 375–381 [Article] [CrossRef] [EDP Sciences] [Google Scholar]
16. HOU Yan, HUANG Kanghuan, ZHANG Yixian, et al.. A multi-objective scheduling optimization for crude oil operations[J]. Industrial Engineering Journal, 2020, 23(4): 131–139 (in Chinese) [Google Scholar]
17. TU Nan, CHANG Liufeng, MEHRBOD Mehrdad, et al.. Multi-objective optimization of closed-loop logistics network with facility expansion[J]. Industrial Engineering Journal, 2013, 16(5): 53–61 (in Chinese) [Google Scholar]
18. ZHANG Lianwen, GUO Haipeng. Introduction to Bayesian networks[M]. Beijing: Science Press, 2006 (in Chinese) [Google Scholar]
19. FRIEDMAN N, GEIGER D, GOLDSZMIDT M. Bayesian network classifiers[J]. Machine Learning, 1997, 29(2/3): 131–163 [Article] [CrossRef] [Google Scholar]
20. MUGHAL M, KIM S. Signal classification and jamming detection in wide-band radios using naive Bayes classifier[J]. IEEE Communications Letters, 2018, 22(7): 1398–1401 [Article] [CrossRef] [Google Scholar]
21. ZHOU Zhihua. Machine Learning[M]. Beijing: Tsinghua University Press, 2016: 1–18 (in Chinese) [Google Scholar]
22. LEE V, LYE D, SUN Y, et al.. Decision tree algorithm in deciding hospitalization for adult patients with dengue haemorrhagic fever in Singapore[J]. Tropical Medicine & International Health, 2010, 14(9): 1154–1159 [Google Scholar]
23. WU Y, YANG H, ZHANG L, et al.. Application of crossover analysis-logistic regression in the assessment of gene-environmental interactions for colorectal cancer[J]. Asian Pacific Journal of Cancer Prevention, 2012, 13(5): 2031–2037 [Article] [Google Scholar]
24. LING Z L, YU K, ZHANG Y W, et al.. Causal learner: a toolbox for causal structure and Markov blanket learning[J]. Pattern Recognition Letters, 2022, 163(2): 92–95 [CrossRef] [Google Scholar]
25. TARVAINEN A, VALPOLa H. Mean teachers are better role models: weight-averaged consistency targets improve semi-supervised deep learning results[C]//31th Conference on Neural Information Processing System, Long Beach, USA, 2017 [Google Scholar]