基于改进极限学习机的水轮机运转状态识别

Abstract
Figure/Table
References (18)
Related Citation (15)

Download: PDF (2116 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract 针对水轮机远程运行状态监测困难的现状，本文提出了基于模拟退火算法的粒子群算法（Simulated annealing algorithm-Particle swarm optimization, SA-PSO）优化极限学习机（Extreme learning machine, ELM）的方式，建立水轮机运转状态识别模型SA-PSO-ELM。对水轮机信号进行互补集合经验模态分解（Complementary ensemble empirical mode decomposition, CEEMD）获得IMF分量，引入皮尔逊相关系数计算各个IMF分量与水轮机信号的相似度，判断信号主导模态和噪声主导模态的分界点，用小波去噪对噪声主导模态降噪，并与其余的IMF分量重构得到去噪信号，同时对去噪后的信号进行分解，计算每个IMF分量的排列熵，构建特征向量。由于SA-PSO精度高不易陷入局部最优的特点和ELM的性能受权值、阈值共同影响的特点，用SA-PSO优化ELM的权值和阈值，构建水轮机运转状态识别模型SA-PSO-ELM。对不同工况下采集的水轮机压力脉动数据，分析了基于PSO-ELM、ELM 及SA-PSO-ELM模型的水轮机运转状态识别正确率、均方误差、决定系数3个指标。结果表明：SA-PSO-ELM更适合于水轮机运转状态识别。
关键词：水轮机；互补集合经验模态分解；粒子群算法；极限学习机；状态识别

Key words：水轮机互补集合经验模态分解粒子群算法极限学习机状态识别

Received: 09 August 2021 Published: 15 January 2022

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	LAN Chaofeng
	SONG Bowen
	GUO Xiaoxia

Cite this article:

LAN Chaofeng,SONG Bowen,GUO Xiaoxia. Identification of hydraulic turbine operation state based on improved limit learning machine[J]. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(1): 132-138.

URL:

http://jvs.sjtu.edu.cn/EN/ OR http://jvs.sjtu.edu.cn/EN/Y2023/V42/I1/132

[1] 孙龙刚,郭鹏程,罗兴锜.水轮机尾水管涡带压力脉动同步及非同步特性研究[J].农业机械学报,2019,50(09):122-129.
Sun Longgang, Guo Pengcheng, Luo Xingkai. Study on synchronous and asynchronous characteristics of pressure pulsation in turbine draft tube vortex belt [J]. Journal of agricultural machinery, 2019,50 (09): 122-129.
[2] 郭雷.水泵水轮机多工况条件下压力脉动研究[D].浙江：浙江大学,2016.
Guo Lei. Study on pressure pulsation of pump turbine under multiple working conditions [D]. Zhejiang: Zhejiang University, 2016.
[3] A. A. Gavrilov,A. V. Sentyabov,K. A. Finnikov. Numerical Investigation of Three-dimensional Cavitating Flow around a Guide Vane of a Water Turbine[J]. Fluid Dynamics,2019,54(4):441-450.
[4] 洪礼聪,王卫玉,陈启卷.基于连续小波变换和卷积神经网络的尾水管涡带状态识别[J].广东电力,2018,31(05):1-6.
Hong Licong, Wang Weiyu, Chen Qijuan. State identification of draft tube vortex zone based on continuous wavelet transform and convolutional neural network [J]. Guangdong electric power, 2018,31 (05): 1-6.
[5] Xin Xia,Wei Ni. A novel failure analysis and diagnosis method for hydraulic-turbine generator unit[J]. Journal of Vibroengineering, 2016, 18(6):3568-3580.
[6] 李水静.基于FOA-PNN的水轮机运转状态识别研究[D].哈尔滨理工大学,2020.
Li Shuijing. Research on turbine operation state identification based on foa-pnn [D]. Harbin University of technology, 2020.
[7] H. V. Gururaja Rao,Nagesh Prabhu,R. C. Mala. Wavelet transform-based protection of transmission line incorporating SSSC with energy storage device[J]. Electrical Engineering,2020(prepublish):1-12.
[8] 王瀚,罗兴锜,薛延刚,丁况.EMD指标能量提取水轮机尾水管动态特征信息[J].水力发电学报,2012,31(05):286-291.
    Wang Han, Luo Xingkai, Xue Yangang, Ding Kuang. Extraction of dynamic characteristic information of hydraulic turbine draft tube by EMD index energy [J]. Journal of hydropower, 2012,31 (05): 286-291.
[9] Yanfei Lu,Rui Xie,Steven Y. Liang. CEEMD-assisted kernel support vector machines for bearing diagnosis[J]. The International Journal of Advanced Manufacturing Technology,2020,106(7):3063-3070.
[10] 孙灵芳,王彤彤,徐曼菲,李霞,朴亨.基于改进CEEMD的薄层污垢超声检测信号去噪[J].仪器仪表学报,2017,38(12):2879-2887.
        Sun Lingfang, Wang Tongtong, Xu Manfei, Li Xia, Pu Heng. Denoising of ultrasonic detection signal of thin layer dirt based on improved ceemd [J]. Journal of instrumentation, 2017,38 (12): 2879-2887.
[11] 朱国俊,陈旻甲,冯建军,刘可然,罗兴锜.基于样本熵的高水头混流式水轮机稳定性分析[J].农业机械学报,2021,52(07):134-140+101.
        Zhu Guojun, Chen Minjia, Feng Jianjun, Liu Keran, Luo Xingmin. Stability analysis of high head Francis Turbine Based on sample entropy [J]. Journal of agricultural machinery, 2021,52 (07): 134-140 + 101.
[12] 詹瀛鱼,程良伦,王涛.解相关多频率经验模态分解的故障诊断性能优化方法[J].振动与冲击,2020,39(01):115-122+149.
         Zhan yingyu, Cheng Lianglun, Wang Tao. Fault diagnosis performance optimization method based on decorrelated multi frequency empirical mode decomposition [J]. Vibration and shock, 2020,39 (01): 115-122 + 149.
[13] Lan Chaofeng,Li Shuijing,Chen Huan,Zhang Wu,Li Hui. Research on running state recognition method of hydro-turbine based on FOA-PNN[J]. Measurement,2021,169.
[14] 朱文龙,周建中,夏鑫,李超顺.基于水电机组运行工况的水轮机压力脉动诊断策略[J].振动与冲击,2015,34(08):26-30+40.
Zhu Wenlong, Zhou Jianzhong, Xia Xin, Li Chaoshun. Pressure fluctuation diagnosis strategy of hydraulic turbine based on operating conditions of hydropower units [J]. Vibration and shock, 2015,34 (08): 26-30 + 40.
[15] Zhihuai Xiao,Xinying He,Xiangqian Fu,O. P. Malik,Yun-Bo Zhao. ACO-Initialized Wavelet Neural Network for Vibration Fault Diagnosis of
Hydroturbine Generating Unit[J]. Mathematical Problems in Engineering,2015,2015.
[16] Zhen Zhao,Jun Zhang,Yigang Sun,Huixin Tian. Fault detection and diagnosis method for batch process based on ELM-based fault feature phase identification[J]. Neural Computing and Applications,2016,27(1):167-173.
[17] 许爱东,李昊飞,程乐峰,余涛.PCA-PSO-ELM配网供电可靠性预测模型[J].哈尔滨工程大学学报,2018,39(06):1116-1122.
    Xu Aidong, Li haofei, Cheng Lefeng, Yu Tao. Pca-pso-elm distribution network power supply reliability prediction model [J]. Journal of Harbin Engineering University, 2018,39 (06): 1116-1122.
[18] W.Li, B.Li, H.L.Guo, Y.X.Fang, F.J.Qiao, S.W.Zhou.The ecg signal classification based on
ensemble learning of pso-elm algorithm[J]. Neural Network World, 2020, 265-279.