基于FWA-DBN的航空发电机偏心故障诊断

doi:10.12305/j.issn.1001-506X.2022.05.39

摘要/Abstract

摘要：

针对具有多并联支路绕组结构的航空发电机在偏心故障下的输出三相电压、电流故障特征差异小, 造成故障不易识别的问题, 提出一种基于烟花算法(fireworks algorithm, FWA)优化深度置信网络(deep belief network, DBN)的故障诊断方法。首先根据有限元法搭建航空发电机模型, 通过仿真获取不同静态、动态偏心故障输出数据; 然后运用FWA训练优化与极限学习机(extreme learning machine, ELM)相结合的DBN网络, 得到最佳DBN-ELM模型结构; 最后由ELM分类器进行故障诊断分类。诊断结果表明, 相较于传统的故障诊断方法, 应用所提方法进行航空发电机偏心故障诊断, 可以获得更高的准确率, 平均准确率达到99.203%。

关键词: 航空发电机, 偏心故障, 烟花算法, 深度置信网络, 极限学习机

Abstract:

A fault diagnosis method based on deep belief network (DBN) optimized by the fireworks algorithm (FWA) is proposed for the problem of small differences in the characteristics of three-phase voltage and current under eccentricity faults in aviation generators with multi-parallel branch winding structures, which are not easy to classify. Firstly, a finite element model of the aviation generator is built and different static and dynamic eccentricity fault data are obtained by simulation. Then the DBN network combining with extreme learning machine (ELM) is optimized by FWA and the optimal DBN-ELM network structure can be obtained. Finally, the fault diagnosis is performed by the ELM classifier. The diagnosis results show that the application of the proposed method for aviation generator eccentricity diagnosis can obtain a higher accuracy with an average accuracy of 99.203% compared with the traditional fault diagnosis methods.

Key words: aviation generator, eccentricity faults, fireworks algorithm (FWA), deep belief network (DBN), extreme learning machine (ELM)

中图分类号:

TM307

杨占刚, 徐海义, 成博源, 石旭东. 基于FWA-DBN的航空发电机偏心故障诊断[J]. 系统工程与电子技术, 2022, 44(5): 1757-1764.

Zhangang YANG, Haiyi XU, Boyuan CHENG, Xudong SHI. Aviation generator eccentricity fault diagnosis based on FWA-DBN[J]. Systems Engineering and Electronics, 2022, 44(5): 1757-1764.

图/表 21

图1

表1

图2

图3

图4

图5

表2

图6

图7

图8

图9

图10

表3

图11

图12

图13

表4

图14

图15

表5

表6

参考文献 24

1	BERSCH K , NUZZO S , CONNOR P H , et al. Thermal and electromagnetic stator vent design optimisation for synchronous generators[J]. IEEE Trans.on Energy Conversion, 2021, 36 (1): 207- 217. doi: 10.1109/TEC.2020.3004393
2	SADEGHI I, EHYA H, FAIZ J. Analytic method for eccentricity fault diagnosis in salient-pole synchronous generators[C]//Proc. of the International Conference on Optimization of Electrical and Electronic Equipment & Intermational Aegean Conference on Electrical Machines and Power Electronics, 2017: 261-267.
3	BRUZZESE C , JOKSIMOVIC G . Harmonic signatures of static eccentricities in the stator voltages and in the rotor current of no-load salient-pole synchronous generators[J]. IEEE Trans.on Industrial Electronics, 2011, 58 (5): 1606- 1624. doi: 10.1109/TIE.2010.2087296
4	GALFARSORO U, MCCLOSKEY A, ZARATE S, et al. Influence of manufacturing tolerances and eccentricities on the unbal anced magnetic pull in permanent magnet synchronous motors[C]//Proc. of the International Conference on Electrical Machines, 2020: 1363-1369.
5	EHYA H , NYSVEEN A , NILSSEN R , et al. Static and dynamic eccentricity fault diagnosis of large salient pole synchronous generators by means of external magnetic field[J]. IET Electric Power Applications, 2021, 15 (7): 890- 902. doi: 10.1049/elp2.12068
6	ILAMPARITHI T C , NANDI S . Identification of spectral components in the line current of eccentric salient pole machines using a binomial series-based inverse air-gap function[J]. IET Electric Power Applications, 2013, 7 (4): 303- 312. doi: 10.1049/iet-epa.2012.0192
7	LASJERDI H , NASIRI-GHEIDARI Z , TOOTOONCHIAN F . Static eccentricity fault diagnosis in wound-rotor resolvers[J]. IEEE Sensors Journal, 2021, 21 (2): 1424- 1432. doi: 10.1109/JSEN.2020.3019260
8	任杰, 王秀和, 赵文良, 等. 永磁同步电机转子偏心空载气隙磁场解析计算[J]. 电机与控制学报, 2020, 24 (8): 26- 32.
	REN J , WANG X H , ZHAO W L , et al. Open circuit magnetic field prediction in permanent magnet synchronous machine with rotor eccentricity[J]. Electric Machines and Control, 2020, 24 (8): 26- 32.
9	崔洪玮. 偏心故障内置式永磁同步电机电磁场分析与诊断方法研究[D]. 哈尔滨工业大学, 2019.
	CUI H W. Research on electromagnetic field analysis and diagnosis method of the interior permanent magnet synchronous motor with eccentricity fault[D]. Harbin: Harbin Institute of Technology, 2019.
10	冯战, 王杰, 黄思思, 等. 基于WP-LSTM的偏心转子马达故障诊断方法[J]. 组合机床与自动化加工技术, 2020, (10): 98- 102, 105.
	FENG Z , WANG J , HUANG S S , et al. Fault diagnosis method for eccentric rotor motor based on WP-LSTM[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2020, (10): 98- 102, 105.
11	任强, 官晟, 王凤军, 等. 基于EEMD和PSO-SVM的电机气隙偏心故障诊断[J]. 组合机床与自动化加工技术, 2021, (2): 73- 76, 85.
	REN Q , GUAN S , WANG F J , et al. Motor air-gap eccentricity fault diagnosis based on EEMD and PSO-SVM[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2021, (2): 73- 76, 85.
12	TAMILSELVAN P , WANG P . Failure diagnosis using deep belief learning based health state classification[J]. Reliability Engineering & System Safety, 2013, 115, 124- 135.
13	李京峰, 陈云翔, 项华春, 等. 基于LSTM-DBN的航空发动机剩余寿命预测[J]. 系统工程与电子技术, 2020, 42 (7): 1637- 1644.
	LI J F , CHEN Y X , XIANG H C , et al. Remaining useful life prediction for aircraft engine based on LSTM-DBN[J]. Systems Engineering and Electronics, 2020, 42 (7): 1637- 1644.
14	马楠, 倪优扬, 葛红娟. 基于DBN的航空发电机故障诊断方法研究[J]. 航空计算技术, 2020, 50 (4): 71- 75.
	MA N , NI Y Y , GE H J . Research on fault diagnosis method of aviation generator based on DBN[J]. Aeronautical Computing Technique, 2020, 50 (4): 71- 75.
15	李俊卿, 陈雅婷, 李斯璇. 基于深度置信网络的同步发电机励磁绕组匝间短路故障预警[J]. 电力自动化设备, 2021, 41 (2): 153- 158.
	LI J Q , CHEN Y T , LI S X . Early warning of inter-turn short circuit fault in excitation windings of synchronous generator based on deep belief network[J]. Electric Power Automation Equipment, 2021, 41 (2): 153- 158.
16	崔江, 郭瑞东, 张卓然, 等. 基于改进DBN的发电机旋转整流器故障特征提取技术[J]. 中国电机工程学报, 2020, 40 (7): 2369- 2376, 2415.
	CUI J , GUO R D , ZHANG Z R , et al. Generator rotating rectifier fault feature extraction technique based on improved DBN[J]. Proceedings of the CSEE, 2020, 40 (7): 2369- 2376, 2415.
17	李益兵, 王磊, 江丽. 基于PSO改进深度置信网络的滚动轴承故障诊断[J]. 振动与冲击, 2020, 39 (5): 89- 96.
	LI Y B , WANG L , JIANG L . Rolling bearing fault diagnosis based on DBN algorithm improved with PSO[J]. Journal of Vibration and Shock, 2020, 39 (5): 89- 96.
18	ZHAO H T , ZHANG C S , NING J X . A best firework updating information guided adaptive fireworks algorithm[J]. Neural Computing & Applications, 2019, 31 (1): 79- 99.
19	CHEN Y G , LI L X , ZHAO X C , et al. Simplified hybrid fireworks algorithm[J]. Knowledge-Based Systems, 2019, 173, 128- 139. doi: 10.1016/j.knosys.2019.02.029
20	WANG W D , LIU K J , YANG C , et al. Cyber physical energy optimization control design for PHEVs based on enhanced firework algorithm[J]. IEEE Trans.on Vehicular Technology, 2021, 70 (1): 282- 291. doi: 10.1109/TVT.2020.3046520
21	TOUFIGHIAN S, FAIZ J, ERFANI-NIK A. Static eccentricity fault detection in salient and non-salient synchronous generators using harmonic components[C]//Proc. of the 12th Power Electronics, Drive Systems, and Technologies Conference, 2021.
22	李海平, 齐卓砾, 胡君朋. 基于FFT-DBN的行星齿轮箱齿面磨损故障智能判定方法研究[J]. 测控技术, 2020, 39 (12): 50- 54, 62.
	LI H P , QI Z L , HU J P . Intelligent judgment of tooth wear fault problems for planetary gearbox based on FFT-DBN[J]. Measurement & Control Technology, 2020, 39 (12): 50- 54, 62.
23	DAI L, PAN P S. Research on intrusion detection based on improved DBN-ELM[C]//Proc. of the International Conference on Communications, Information System and Computer Engineering, 2019: 495-499.
24	XIE Y C , ZOU J X , LI Z L , et al. A novel deep belief network and extreme learning machine based performance degradation prediction method for proton exchange membrane fuel cell[J]. IEEE Access, 2020, 8, 176661- 176675. doi: 10.1109/ACCESS.2020.3026487

结构	参数
额定功率S/kVA	65
功率因数λ	0.95
额定电压U/V	407
频率f/Hz	400
定子槽数	54
转子极对数p	3
定子外径d_o/mm	225
定子内径d_i/mm	160
最小气隙d_r/mm	1
并联支路数a	2

故障类别	e_A/V	e_B/V	e_C/V
正常	332.46	332.44	332.44
ρ_S=0.25	333.17	332.84	333.27
ρ_S=0.50	332.95	332.97	333.02
ρ_S=0.75	333.11	333.46	333.09
ρ_D=0.25	332.32	332.73	332.63
ρ_D=0.50	332.73	332.89	332.38
ρ_D=0.75	332.81	333.41	332.46

故障类别	U_A/V	U_B/V	U_C/V	i_A/A	i_B/A	i_C/A
正常	330.881	330.032	330.692	120.320	120.012	120.252
ρ_S=0.25	330.784	330.070	330.577	120.285	120.025	120.210
ρ_S=0.50	331.445	330.845	331.108	120.525	120.307	120.403
ρ_S=0.75	332.356	331.784	332.467	120.857	120.649	120.897
ρ_D=0.25	330.708	329.918	330.124	120.257	119.970	120.045
ρ_D=0.50	333.391	333.164	333.442	121.233	121.151	121.252
ρ_D=0.75	332.364	331.583	331.621	120.860	120.576	120.589

参数	取值
迭代次数	200
烟花数量n	50
爆炸半径控制因子α	4
爆炸数量控制因子β	5
爆炸上界x_UB	500
爆炸下界x_LB	1

序号	RBM网络结构	测试集/%	训练集/%	训练时长/s
1	1000-500-250-100	95.233	98.661	46.56
2	800-400-200-100	94.281	96.852	37.44
3	500-250-100-50	83.995	86.997	23.52
4	本文	98.856	99.203	8.16