Abstract:In order to explore the influence of assembly parameters of aero-engine high-pressure rotor assembly on the vibration response of the whole machine, the mechanism relationship between geometric deviation, unbalance deviation and vibration mode in assembly stage is described by unbalanced response vibration equation, and an improved model, MC-XGBoost prediction model, based on maximum correlation (MC) coefficient and extreme gradient lifting (XGBoost) is proposed. The key parameters that affect the vibration performance, namely the most relevant geometric / unbalanced assembly parameters, are screened by the MC coefficient, and then the selected assembly parameters are brought into the XGBoost model to predict the vibration characteristics. The prediction model is verified by experimental data, and the results show that the proposed MC-XGBoost prediction model has higher prediction accuracy than RF and GBDT algorithm models, and can provide an effective evaluation method for aero-engine assembly quality-oriented vibration characteristic evaluation.
Key words: high-pressure rotor; assembly quality; vibration characteristics; maximum correlation coefficient; XGBoost model
梅潇,池华山,岳聪,范建瑜,刘宗沁. 基于MC-XGBoost模型的航空发动机振动特性预测[J]. 振动与冲击, 2022, 41(16): 271-277.
MEI Xiao, CHI Huashan, YUE Cong, FAN Jianyu, LIU Zongqin. Prediction of aeroengine vibration characteristics based on an MC-XGBoost model. JOURNAL OF VIBRATION AND SHOCK, 2022, 41(16): 271-277.
[1] 张子豪,郭俊康,洪军,等. 航空发动机高压转子装配偏心预测和相位优化的智能算法应用研究[J]. 西安交通大学学报,2021,55(2):603-609
ZHANG Zihao, GUO Junkang, HONG Jun, et al. Application study of intelligent algorithms for prediction and phase optimization of assembly eccentricity of aero-engine high pressure rotor[J]. Journal of Xi’an Jiaotong University,2021,55(2):603-609.
[2] 岳 聪,任兴民,邓旺群,等. 航空发动机转子突加不平衡参数分析及LQR 控制技术应用[J]. 振动与冲击,2015,34(17):175-179
YUE Cong, REN Xingming, DENG Wangqun, et al. Parametric analysis due to sudden unbalance change and LQR control application for aeroengine rotor systems[J]. Journal of Vibration and Shock,2015,34(17):175-179.
[3] Zhang Maowei, Liu Yongmeng, Sun Chuanzhi, et al. Measurements error propagation and its sensitivity analysis in the aero-engine multistage rotor assembling process[J]. Review of Scientific Instruments, 2019,90(11):1-11
[4] 孟亮国. 航空发动机转子装配工艺优化方法研究[D].大连理工大学,2019.
MENG Liangguo. Research on process optimization of aeroengine rotor assembly[D]. Dalian University of Technology,2019.
[5] 刘鑫. 航空发动机转子装配精度预测及堆叠[D]. 大连理工大学,2019.
LIU Xin. Assembly accuracy prediction and stacking of aeroengine rotor[D]. Dalian University of Technology,2019.
[6] 余坚,谢寿生,任立通.拉杆转子装配振动检测分形研究[J]. 振动与冲击,2014,33 (14)84–88.
YU Jian, XIE Shousheng, Ren Litong,et al.Fractal research on theassembly vibration detection of rod fastening rotor[ J]. Journal of Vibration and Shock,2014,33 (14)84–88.
[7] Westreich D, Lessler J, Funk J. Propensity score esti-mation:neura lnetworks, support vector machines,decision tree(CART),and meta-classifiers as alternatives to logistic regression[J]. Journal of Clinical Epidemiology,2010,63(8):826-33.
[8] Kacar K, Rocca M A, Copetti M,et al. Overcoming the clinical-MR imaging paradox of multiple sclerosis: MR imaging data assessed with a random forest approach[J]. Journal of Neuroradiology, 2011, 32 (11) 2098-2102
[9] Friedman J H. Greedy function approximation: a gradient boosting machine[J]. Annals of Statistics, 2001, 29(5):1189-1232.
[10] Chen T , Guestrin C . XGBoost: a scalable tree boosting system[C]// the 22nd ACM SIGKDD International Conference. ACM, 2016: 785-794.
[11] 李卫星,陶建峰,贾程锦,等. 同步压缩小波与极限梯度提升树融合的柴油机失火故障诊断[J]. 西安交通大学学报,2019,53(2):48-54
LI Weixing, TAO Jianfeng, JIA Chengjin, et al. A diagnostic method for diesel engine misfire based on integrating of synchro-squeezed wavelet transform xgboost[J]. Journal of Xi’an Jiaotong University,2019,53(2):48-54.
[12] 钱力扬.基于随机森林和XGBoost的大型风力机故障诊断方法研究[D]. 浙江大学,2018.
QIAN Liyang.A data-driven design for fault detection of wind turbines using random forests ang xgboost[D]. Zhejiang University,2018.
[13] Debaditya C, Hazem E.Early detection of faults in HVAC systems using an XGBoost model with a dynamic threshold[J]. Energy & Buildings,2019(185)326-344
[14] 张荣涛,陈志高,李彬彬,等.基于深度卷积神经网络模型和XGBoost 算法的齿轮箱故障诊断研究[J]. 机械强度,2020,42( 5) : 1059-1066
ZHANG Rongtao, CHEN Zhigao, LI Binbin, el al. Reserach on gear box fault diagnosis based on DCNN and XGBoost algorithm[J]. Journal of Mechanical Strength. 2020,42( 5) : 1059-1066.
[15] 王伟,匡祯辉,谢少捷,等. 热镀锌钢卷力学性能GBDT预报模型[J].福州大学学报(自然科学版),2020,48,(4):603-609
WANG wei,KUANG Zhenhui,XIE Shaojie ,et al.Research on prediction model of hot dip galvanized steel coils[J]. Journal of Fuzhou University(Natural Science Edition) ,2020,48,(4):603-609.
[16] 陈振宇,刘金波,李晨,等. 基于 LSTM 与 XGBoost 组合模型的超短期电力负荷预测[J].电网技术,2020,44(2):615-620
CHEN Zhenyu, LIU Jinbo, LI Chen, et al. Ultra short-term power load forecasting based on combined LSTM-XGBoost model[J]. Power System Technology,2020,44(2):615-620.
[17] Breiman L ,Friedman J H. Estimating optimal transformations for multiple-regression and correlation[J]. Journal of the American Statistical Association, 1985,80(391):580–598