Fault diagnosis of synchronous generator rotating rectifier based on CEEMD and improved ELM
-
摘要:
针对目前应用于航空发电机旋转整流器故障诊断中的人工智能算法存在诊断速度慢、参数选取困难等问题,提出一种基于互补式集合经验模态分解(CEEMD)与樽海鞘优化的极限学习机(SSA-ELM)故障诊断方法。在有限元软件Maxwell与Simplorer中搭建三级式电机模型,采集励磁电流信号,利用CEEMD将励磁电流信号分解为一系列模态分量,构建故障特征参量,再通过樽海鞘群算法(SSA)优化极限学习机的训练参数
$ \omega $ 和$ b $ ,并对故障进行诊断,最后通过实验平台验证所提方法。结果证明了三级式同步电机有限元模型的有效性,所提方法相校于现有方法,具有更高的故障诊断准确率与分类速度。Abstract:Aiming at the problems of slow diagnostic speed and difficulty in pardmeter selection in artifical intelligence algorithms currently used in fault diagnosis of aviation generator rotating rectifiers, an extreme learning machine (SSA-ELM) and complementary ensemble empirical mode decomposition (CEEMD) based fault diagnosis approach is investigated. In the finite element software Maxwell and Simplorer, the three-stage motor model is built, the excitation current signal is collected, the excitation current signal is decomposed into a series of modal components by CEEMD, and the fault characteristic parameters are constructed. Then the training parameters and parameters of the extreme learning machine are optimized by the SSA algorithm, and the fault is diagnosed. Finally, it is verified by the experimental platform. The results show the effectiveness of the three-stage synchronous motor finite element model.Compared with the existing methods, the rotating rectifier diode fault diagnosis method based on CEEMD and SSA-ELM has higher fault diagnosis accuracy and classification speed.
-
表 1 主发电机基本参数
Table 1. Basic parameters of main generator
额定功率/kVA 额定电压/V 额定频率/Hz 极对数 40 115 400 2 表 2 二极管开路故障下的5类故障状态
Table 2. Five kinds of fault states under diode open circuit fault
故障类型 状态描述 故障二极管 类别Ⅰ 二极管正常 类别Ⅱ 单个二极管开路 D1、D2、D3、D4、D5、D6 类别Ⅲ 同桥臂双二极管开路 D1D6、D2D5、D3D4 类别Ⅳ 不同桥臂双二极管开路
(同为上桥臂或下桥臂)D1D2、D1D3、D2D3、D6D5、D6D4、D5D4 类别Ⅴ 不同桥臂双二极管开路
(一个为上桥臂一个为下桥臂)D1D5、D1D4、D2D6、D2D4、D3D6、D3D5 表 3 不同负载条件下各分类方法仿真结果性能对比
Table 3. Simulation result of performance comparison of different classification methods under different load conditions
负载条件 训练时间/s 测试时间/s 本文方法 ELM PSO-ELM SVM SSA-SVM 本文方法 ELM PSO-ELM SVM SSA-SVM 空载 10.2598 0.0128 100 0 0.0129 0.0127 84.00 10.9653 14.6525 0.0127 1.5 kW负载 10.3564 0.0135 100 0 0.0145 0.0136 86.00 11.3256 14.7869 0.0145 3 kW负载 10.3256 0.0139 100 0 0.0142 0.0137 87.60 10.2569 14.6598 0.0144 混合负载 29.8579 0.0405 100 0 0.0435 0.0412 81.20 32.1546 44.2698 0.0412 负载条件 准确率/% 诊断方差 本文方法 ELM PSO-ELM SVM SSA-SVM 本文方法 ELM PSO-ELM SVM SSA-SVM 空载 95.60 2.5689 82.5320 0.0232 91.60 0.2365 20.3012 0.0362 100 0 1.5 kW负载 96.40 3.2564 83.2365 0.0210 94.00 0.2589 20.2562 0.0359 100 0 3 kW负载 95.60 2.3546 82.4562 0.0234 93.20 0.3545 20.2564 0.0362 100 0 混合负载 95.60 8.2546 247.6020 0.0694 89.6 0.8456 60.6526 0.1089 100 0 表 4 不同负载条件下各分类方法实验结果性能对比
Table 4. Experimental results of performance comparison of different classification methods under different load conditions
负载条件 训练时间/s 测试时间/s 本文方法 ELM PSO-ELM SVM SSA-SVM 本文方法 ELM PSO-ELM SVM SSA-SVM 空载 10.2598 0.0128 100 0 0.0129 0.0127 84.00 10.2489 14.6525 0.0127 1.5 kW负载 10.3564 0.0135 100 0 0.0145 0.0136 86.00 10.2587 14.7869 0.0145 3 kW负载 10.3256 0.0139 100 0 0.0142 0.0137 87.60 10.8596 14.6598 0.0144 混合负载 29.8579 0.0405 100 0 0.0435 0.0412 81.20 30.2156 44.2698 0.0412 负载条件 准确率/% 诊断方差 本文方法 ELM PSO-ELM SVM SSA-SVM 本文方法 ELM PSO-ELM SVM SSA-SVM 空载 95.60 2.3568 82.5320 0.0232 91.60 0.2156 20.3012 0.0362 100 0 1.5 kW负载 96.40 2.9586 83.2365 0.0210 94.00 0.2698 20.2562 0.0359 100 0 3 kW负载 95.60 2.4788 82.4562 0.0234 93.20 0.2568 20.2564 0.0362 100 0 混合负载 95.60 7.8964 247.6020 0.0694 89.6 0.7549 60.6526 0.1089 100 0 -
[1] 孔祥浩, 张卓然, 陆嘉伟, 等. 分布式电推进飞机电力系统研究综述[J]. 航空学报, 2018, 39(1): 021651.KONG X H, ZHANG Z R, LU J W, et al. Review of electric power system of distributed electric propulsion aircraft[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(1): 021651(in Chinese). [2] MADONNA V, GIANGRANDE P, GALEA M. Electrical power generation in aircraft: Review, challenges, and opportunities[J]. IEEE Transactions on Transportation Electrification, 2018, 4(3): 646-659. doi: 10.1109/TTE.2018.2834142 [3] 唐军祥. 航空三级式发电机旋转整流器故障诊断技术研究[D]. 南京: 南京航空航天大学, 2017: 1-3.TANG J X. Research on fault diagnosis technology of aircraft three-stage generator rotating rectifier[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2017: 1-3(in Chinese) . [4] BATZEL T D, SWANSON D C. Prognostic health management of aircraft power generators[J]. IEEE Transactions on Aerospace and Electronic Systems, 2009, 45(2): 473-482. doi: 10.1109/TAES.2009.5089535 [5] 任磊, 韦徵, 龚春英, 等. 电力电子电路功率器件故障特征参数提取技术综述[J]. 中国电机工程学报, 2015, 35(12): 3089-3101. doi: 10.13334/j.0258-8013.pcsee.2015.12.021REN L, WEI Z, GONG C Y, et al. Fault feature extraction techniques for power devices in power electronic converters: A review[J]. Proceedings of the CSEE, 2015, 35(12): 3089-3101(in Chinese). doi: 10.13334/j.0258-8013.pcsee.2015.12.021 [6] TANTAWY A, KOUTSOUKOS X, BISWAS G. Aircraft power generators: Hybrid modeling and simulation for fault detection[J]. IEEE Transactions on Aerospace and Electronic Systems, 2012, 48(1): 552-571. doi: 10.1109/TAES.2012.6129655 [7] BATZEL T D, SWANSON D C, DEFENBAUGH J F. Predictive diagnostics for the main field winding and rotating rectifier assembly in the brushless synchronous generator[C]//4th IEEE International Symposium on Diagnostics for Electric Machines, Power Electronics and Drives. Piscataway: IEEE Press, 2003: 349-354. [8] 叶纪青. 航空发电机旋转整流器故障特征提取技术研究[D]. 南京: 南京航空航天大学, 2017: 29-40.YE J Q. Research on feature extraction method for rotating rectifier of aircraft generator[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2017: 29-40(in Chinese). [9] 张敬, 李颖晖, 朱喜华, 等. 三级式发电机旋转整流器故障特征提取[J]. 微电机, 2011, 44(7): 92-96. doi: 10.3969/j.issn.1001-6848.2011.07.022ZHANG J, LI Y H, ZHU X H, et al. Fault feature extraction for rotating rectifie of three-stage generator[J]. Micromotors, 2011, 44(7): 92-96(in Chinese). doi: 10.3969/j.issn.1001-6848.2011.07.022 [10] 刘勇智, 刘聪. 基于EMD和LS-SVM的旋转整流器故障诊断方法研究[J]. 微电机, 2012, 45(4): 21-24. doi: 10.3969/j.issn.1001-6848.2012.04.006LIU Y Z, LIU C. Fault diagnose of rotating rectifier based on EMD and LS-SVM[J]. Micromotors, 2012, 45(4): 21-24(in Chinese). doi: 10.3969/j.issn.1001-6848.2012.04.006 [11] 王潇雅, 崔江, 唐军祥, 等. 基于极限学习机的航空旋转整流器故障诊断技术研究[J]. 机械制造与自动化, 2017, 46(5): 219-222. doi: 10.19344/j.cnki.issn1671-5276.2017.05.056WANG X Y, CUI J, TANG J X, et al. Fault diagnosis technology research of aircraft generator rotating rectifier based on extreme learning machine[J]. Machine Building & Automation, 2017, 46(5): 219-222(in Chinese). doi: 10.19344/j.cnki.issn1671-5276.2017.05.056 [12] 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. doi: 10.13334/j.0258-8013.pcsee.190858CUI 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(in Chinese). doi: 10.13334/j.0258-8013.pcsee.190858 [13] 崔江, 冯赛, 张卓然, 等. 基于BLS的无刷发电机旋转整流器特征提取技术研究[J]. 中国电机工程学报, 2020, 40(12): 4004-4013. doi: 10.13334/J.0258-8013.PCSEE.190983CUI J, FENG S, ZHANG Z R, et al. Research on feature extraction technology of brushless generator rotating rectifier based on BLS[J]. Proceedings of the CSEE, 2020, 40(12): 4004-4013(in Chinese). doi: 10.13334/J.0258-8013.PCSEE.190983 [14] 周小麟, 童晓阳. 基于CEEMD-SBO-LSSVR的超短期风电功率组合预测[J]. 电网技术, 2021, 45(3): 855-864. doi: 10.13335/j.1000-3673.pst.2020.0584ZHOU X L, TONG X Y. Ultra-short-term wind power combined prediction based on CEEMD-SBO-LSSVR[J]. Power System Technology, 2021, 45(3): 855-864(in Chinese). doi: 10.13335/j.1000-3673.pst.2020.0584 [15] 韩宏泉, 吴珊, 侯本伟. 采用核极限学习机的短期需水量预测模型[J]. 哈尔滨工业大学学报, 2022, 54(2): 17-24. doi: 10.11918/202012021HAN H Q, WU S, HOU B W. Short-term water demand prediction model using kernel-based extreme learning machine[J]. Journal of Harbin Institute of Technology, 2022, 54(2): 17-24(in Chinese). doi: 10.11918/202012021 -