-
摘要:
为了提高航空发动机工作状态识别准确率和效率,避免人工识别中存在的误判和耗时耗力问题,提出了基于混沌脉冲蝙蝠算法(CRBA)优化的多核支持向量数据描述(CRBA-MKSVDD)智能识别方法。研究了多核支持向量数据描述(MKSVDD)改进策略,引入混沌脉冲发射率提高了蝙蝠算法(BA)的收敛速度和收敛精度,得到了CRBA;通过CRBA优化MKSVDD的惩罚因子和核参数,同时对飞参数据进行了特征提取;基于特征飞参数据训练了CRBA-MKSVDD分类器,并对某型发动机一个飞行架次的工作状态进行了识别。结果表明,该方法识别准确率达到97.547 9%,可用于与发动机工作状态的相关研究和应用。
-
关键词:
- 多核支持向量数据描述(MKSVDD) /
- 改进蝙蝠算法 /
- 航空发动机 /
- 工作状态识别 /
- 飞参数据
Abstract:In order to ameliorate the accuracy and efficiency of aero-engine working condition identification, and to avoid the misjudgment and time-consuming problems in manual identification of aero-engine working condition, an intelligent recognition method, multi-kernel support vector data description based on chaotic rate bat algorithm (CRBA-MKSVDD), is proposed. The improved strategy of multi-kernel support vector data description (MKSVDD) is researched. The chaotic rate method is introduced to improve the convergence speed and convergence accuracy of the bat algorithm (BA), and the chaotic rate bat algorithm (CRBA) is obtained with this method. The penalty factor and kernel parameter of MKSVDD are optimized by CRBA and the characteristics of the flight parameters have been extracted. The CRBA-MKSVDD classifiers are trained based on the characteristics of flight parameters, and the working condition of a certain type of aero-engine in one sortie is identified by the proposed method. The results show that the accuracy of aero-engine working condition identified by the proposed method is 97.547 9%, which means that the method can be used in the research and application related to aero-engine working condition.
-
表 1 某型航空发动机飞参数据格式
Table 1. Flight parameter format of a type of aero-engine
参数 名称 记录时间/0.1s 记录时长 N1/% 低压转速 N2/% 高压转速 T6/℃ 涡轮后燃气温度 A1/(°) 风扇进口可调叶片角度 T1/K 发动机进口总温 T25/K 风扇内涵出口总温 P31/kPa 压气机出口压力 P6/kPa 涡轮后出口压力 PH/kPa 发动机舱压 PLA/(°) 油门杆位置 H/km 飞行高度 A2/(°) 压气机进口可调静子叶片角度 W/kg 主燃油流量给定值 A8/cm2 喷口面积 表 2 分类器测试结果
Table 2. Test results of classifiers
分类器 惩罚因子 核参数 识别准确率/% 测试时间/s 慢车 节流 中间及以上 最大 慢车 节流 中间及以上 最大 慢车 节流 中间及以上 最大 慢车 节流 中间及以上 最大 CV-SVDD 0.7933 0.7827 0.9926 0.0447 221.07 322.26 91.07 100.70 95.91 94.29 81.41 84.37 29.85 39.14 15.75 14.20 CV-MKSVDD 0.022 2 0.4878 0.5149 0.9930 576.72 677.08 827.71 161.73 97.04 96.16 83.57 90.22 37.93 45.71 16.26 16.51 BA-SVDD 0.1531 0.0100 0.5245 0.5832 561.73 656.93 280.13 327.91 95.66 96.55 84.65 86.14 16.08 20.06 8.67 5.40 BA-MKSVDD 0.2608 0.4970 0.5512 1.000 481.60 595.19 499.96 520.38 98.00 94.42 88.31 92.74 19.46 27.06 12.65 8.64 CRBA-MKSVDD 0.1956 0.3949 0.7397 0.3726 463.24 638.42 440.23 957.35 98.67 97.93 92.05 93.36 18.27 25.31 10.15 7.66 -
[1] 何立明.飞机推进系统原理[M].北京:国防工业出版社, 2006:236-237.HE L M.Aircraftpropulsion system principles[M]. Beijing:National Defence of Industry Press, 2006:236-237(in Chinese). [2] SIMON D L, LITT J S.A data filter for identifying steady-state operating points in engine flight data for condition monitoring applications[J]. Journal of Engineering for Gas Turbines and Power, 2011, 133(7):1-8. [3] TAX D M J, DUIN R P W.Support vector data description[J]. Machine Learning, 2004, 54(1):45-66. doi: 10.1023/B:MACH.0000008084.60811.49 [4] 丁昊, 李旭东, 赵慧洁.基于SVDD的三维目标多视点视图建模[J].北京航空航天大学学报, 2012, 38(11):1517-1521. http://bhxb.buaa.edu.cn/CN/abstract/abstract12456.shtmlDING H, LI X D, ZHAO H J.Method of multi-view modeling for 3D target based on SVDD[J]. Journal of Beijing University of Aeronautics and Astronautics, 2012, 38(11):1517-1521(in Chinese). http://bhxb.buaa.edu.cn/CN/abstract/abstract12456.shtml [5] 唐明珠, 王岳斌, 阳春华.一种改进的支持向量数据描述故障诊断方法[J].控制与决策, 2011, 26(7):967-972. http://d.old.wanfangdata.com.cn/Periodical/kzyjc201107002TANG M Z, WANG Y B, YANG C H.Modified support vector data description for fault diagnosis[J]. Control and Decision, 2011, 26(7):967-972(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/kzyjc201107002 [6] 唐炬, 林俊亦, 卓然, 等.基于支持向量数据描述的局部放电类型识别[J].高电压技术, 2013, 39(5):1046-1053. doi: 10.3969/j.issn.1003-6520.2013.05.004TANG J, LIN J Y, ZHUO R, et al.Partial discharge type recognition based on support vector data description[J]. High Voltage Engineering, 2013, 39(5):1046-1053(in Chinese). doi: 10.3969/j.issn.1003-6520.2013.05.004 [7] PENG X J, XU D.Efficient support vector data descriptions for novelty detection[J]. Neural Computing & Applications, 2012, 21(8):2023-2032. [8] 曲建岭, 李晓娟, 司敬国, 等.基于飞参数据的发动机工作状态识别方法研究[C]//第32届中国控制会议.西安: 西北工业大学, 2013: 3565-3569.QU J L, LI X J, SI J G, et al.On recognitionmethod of aeroengine working conditions based on flight data[C]//Proceedings of the 32nd Chinese Control Conference.Xi'an: Northwestern Polytechnical University, 2013: 3565-3569(in Chinese). [9] 周胜明, 曲建岭, 高峰, 等.基于HE-SVDD的航空发动机工作状态识别[J].仪器仪表学报, 2016, 37(2):308-315. doi: 10.3969/j.issn.0254-3087.2016.02.010ZHOU S M, QU J L, GAO F, et al.Aero-engine working condition recognition based on HE-SVDD[J]. Chinese Journal of Scientific Instrument, 2016, 37(2):308-315(in Chinese). doi: 10.3969/j.issn.0254-3087.2016.02.010 [10] ASHKEZARI A D, MA H, SAHA T K, et al.Application of fuzzy support vector machine for determining the health index of the insulation system of in-service power transformers[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2013, 20(3):965-972. doi: 10.1109/TDEI.2013.6518966 [11] JORDI MUNOZ-MARI L B, CAMPS-VAILS G.A support vector domain description approach to supervised classification of remote sensing images[J]. Transactions on Geoscience and Remote Sensing, 2007, 45(8):2683-2692. doi: 10.1109/TGRS.2007.897425 [12] 卢明, 刘黎辉, 吴亮红.多核支持向量数据描述分类方法研究[J].计算机工程与应用, 2016, 52(18):68-73. doi: 10.3778/j.issn.1002-8331.1412-0205LU M, LIU L H, WU L H.Research on multi-kernel support vector data description method of classification[J]. Computer Engineering and Applications, 2016, 52(18):68-73(in Chinese). doi: 10.3778/j.issn.1002-8331.1412-0205 [13] 刘明.支持向量机中Sigmoid核函数的研究[D].西安: 西安电子科技大学, 2010: 34-35. http://cdmd.cnki.com.cn/Article/CDMD-10701-2010083361.htmLIU M.The study on Sigmoid kernel function in support vector machine[D]. Xi'an: Xidian University, 2010: 34-35(in Chinese). http://cdmd.cnki.com.cn/Article/CDMD-10701-2010083361.htm [14] ZHOU S S, LIU H W, YE F.Variant of Gaussian kernel and parameter setting method for nonlinear SVM[J]. Neurocomputing, 2009, 72(13):2931-2937. doi: 10.1016-j.neucom.2008.07.016/ [15] YANG X S.A new metaheuristic bat-inspired algorithm[C]//Nature Inspired Cooperative Strategies for Optimization (NISCO 2010).Berlin: Springer-Verlag, 2010, 284: 65-74. [16] YANG X S.Nature inspired meta-heuristic algorithms[M]. 2nd ed.Frome:Luniver Press, 2010:97-104. [17] 李枝勇, 马良, 张惠珍.蝙蝠算法收敛性分析[J].数学的实践与认识, 2013, 43(12):182-190. doi: 10.3969/j.issn.1000-0984.2013.12.026LI Z Y, MA L, ZHANG H Z.Convergenceanalysis of bat algorithm[J]. Mathematics in Practice and Theory, 2013, 43(12):182-190(in Chinese). doi: 10.3969/j.issn.1000-0984.2013.12.026 [18] ADARSH B R, RAGHUNATHAN T, JAYABARATHI T, et al.Economic dispatch using chaotic bat algorithm[J]. Energy, 2016, 96(2):666-675. [19] KARRI C, JENA U.Fast vector quantization using a bat algorithm for image compression[J]. Engineering Science & Technology, an International Journal, 2016, 19(2):769-781. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=Doaj000004650871 [20] RAHIMI A, BAVAFA F, AGHABABAEI S, et al.The online parameter identification of chaotic behaviour in permanent magnet synchronous motor by self-adaptive learning bat-inspired algorithm[J]. International Journal of Electrical Power & Energy Systems, 2016, 78(6):285-291. [21] 李煜, 裴宇航, 刘景森.融合均匀变异与高斯变异的蝙蝠优化算法[J].控制与决策, 2017, 32(10):1775-1781. http://d.old.wanfangdata.com.cn/Periodical/kzyjc201710006LI Y, PEI Y H, LIU J S.Bat optimal algorithm combined uniform mutation with Gaussian mutation[J]. Control and Decision, 2017, 32(10):1775-1781(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/kzyjc201710006 [22] 刘长平, 叶春明.具有混沌搜索策略的蝙蝠优化算法及性能仿真[J].系统仿真学报, 2013, 25(6):1183-1195. http://d.old.wanfangdata.com.cn/Periodical/xtfzxb201306008LIU C P, YE C M.Batalgorithm with chaotic search strategy and analysis of its property[J]. Journal of System Simulation, 2013, 25(6):1183-1195(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/xtfzxb201306008 [23] JAVAD H, REZA S, NEDA N.Weighted support vector data description based on chaotic bat algorithm[J]. Applied Soft Computing, 2017, 60:540-551. doi: 10.1016/j.asoc.2017.07.038 [24] GANDOM A H, YANG X S.Chaotic bat algorithm[J]. Journal of Computational Science, 2014, 5(2):224-232. doi: 10.1016/j.jocs.2013.10.002