Cavitation State Recognition Method of Hydraulic Turbine Based on Energy-Entropy Features and Improved Stacked Denoising Auto Encoder

doi:10.12096/j.2096-4528.pgt.260116

Abstract

Abstract:

Objectives Aiming at the problem that the acoustic emission (AE) signal induced by the cavitation of a hydraulic turbine is disturbed by background noise and the fault is difficult to identify, a state recognition method based on energy-entropy features and Harris hawks optimization (HHO) algorithm combined with 3-fold cross-validation (3Fold) to optimize stacked denoising auto encoder (SDAE) is proposed. Methods First, the variational mode decomposition algorithm is used to decompose the signal, and then a series of intrinsic mode functions are obtained. Second, the energy and entropy features of the two intrinsic mode functions are extracted, which have the largest correlation coefficient with the original signal. A 12-dimensional feature vector is constructed and then the vector is imported into the recognition model. Third, the HHO algorithm combined with 3Fold is used to optimize the hyper-parameters of SDAE. Finally, the optimal parameters obtained by the HHO-3Fold-SDAE algorithm and other algorithms are input into the model for comparison. Results Compared with other algorithms, the HHO-3Fold-SDAE algorithm has smaller loss rate and accuracy variance, and higher average accuracy. Compared with SDAE, its test set average accuracy is increased by 6%. Compared with HHO-SDAE, its test set average accuracy is increased by 4% and accuracy variance is decreased by 17%. Conclusions The proposed method can be used to classify and recognize AE signals induced by cavitation of hydraulic turbine, and provide reference for the condition monitoring of hydraulic machinery.

Key words: hydropower, hydraulic turbine, cavitation state recognition, Harris hawks optimization (HHO) algorithm, stacked denoising auto encoder (SDAE), entropy

CLC Number:

TK 73

Zhen LIU, Zhong LIU, Shuyun ZOU, Zehua ZHOU, Shuaicheng QIAO. Cavitation State Recognition Method of Hydraulic Turbine Based on Energy-Entropy Features and Improved Stacked Denoising Auto Encoder[J]. Power Generation Technology, 2026, 47(1): 176-184.

Figures/Tables 14

Fig. 1 Feature extraction process of AE signal induced by cavitation of hydraulic turbine

Fig. 2 Hyperparameter optimization process of HHO-3Fold-SDAE

Fig. 3 Cavitation recognition model for hydraulic turbine

Tab. 1 Sample division of AE signal induced by cavitation of hydraulic turbine

编号	空化状态	空化系数 $σ$	训练集	测试集
1	无空化	$σ i < σ$	210	90
2	初生空化	$σ i$	210	90
3	空化发展	$σ 1 < σ < σ i$	210	90
4	临界空化	$σ 1$	210	90

Tab. 1 Sample division of AE signal induced by cavitation of hydraulic turbine

编号	空化状态	空化系数 $σ$	训练集	测试集
1	无空化	$σ i < σ$	210	90
2	初生空化	$σ i$	210	90
3	空化发展	$σ 1 < σ < σ i$	210	90
4	临界空化	$σ 1$	210	90

Tab. 2 Five IMFs with the largest correlation coefficients

分量	IMF1	IMF2	IMF3	IMF4	IMF5
相关系数	0.702	0.497	0.402	0.274	0.265

Fig. 4 Time domain diagram of two IMFs with the largest correlation coefficients

Tab. 3 Parameter initialization of HHO-3Fold-SDAE

参数	数值
隐含层数	3
激活函数	Sigmoid
批量数	84
循环次数	200
HHO种群个数H	5
HHO迭代次数T	30

Tab. 4 Optimized results of HHO-3Fold-SDAE

参数	取值
第1个隐藏层节点数H_i₁	43
第2个隐藏层节点数H_i₂	71
第3个隐藏层节点数H_i₃	62
反向传播学习率H_i₄	2
反向传播循环次数H_i₅	2 836

Tab. 5 Comparison of recognition results of different models

模型	训练集平均准确率/%	训练集准确率方差	测试集平均准确率/%	测试集准确率方差	平均总时长/s
HHO-3Fold-SDAE	98.45	0.78	92.75	1.06	167.99
PSO-3Fold-SDAE	98.02	1.03	90.22	1.05	89.71
HHO-SDAE	95.42	0.87	89.19	1.28	124.31
PSO-SDAE	95.19	1.69	89.72	1.14	78.06
SDAE	87.77	0.93	87.28	1.08	59.15

Fig. 5 Classification results of HHO-3Fold-SDAE training set

Fig. 6 Classification results of HHO-3Fold-SDAE test set

Fig. 7 Confusion matrix of HHO-3Fold-SDAE training set

Fig. 8 Confusion matrix of HHO-3Fold-SDAE test set

Fig. 9 Convergence curves of loss rate and accuracy rate for different models

References 24

[1]	李书勇，蔡海青，沈娜，等．不同类型新能源接入对微电网频率的影响[J]．分布式能源，2024，9(2)：8-18．
	LI S Y， CAI H Q， SHEN N，et al ．Impact of different types of new energy access on microgrid frequency[J]．Distributed Energy，2024，9(2)：8-18．
[2]	FENG J， LIU B， LUO X，et al ．Experimental investigation on characteristics of cavitation-induced vibration on the runner of a bulb turbine[J]．Mechanical Systems and Signal Processing，2023，189：110097． doi:10.1016/j.ymssp.2023.110097
[3]	王照伟，郭雯君，宋向金，等．融合TKEO和Goerztel算法的感应电机转子断条故障诊断[J]．电工技术学报，2024，39(12)：3679-3690．
	WANG Z W， GUO W J， SONG X J，et al ．Fault diagnosis of broken rotor bar in induction motor based on TKEO and goertzel algorithm[J]．Transactions of China Electrotechnical Society，2024，39(12)：3679-3690．
[4]	刘忠，王文豪，邹淑云，等．水轮机空化声发射信号的联合降噪与特征提取[J]．水力发电学报，2022，41(12)：145-152．
	LIU Z， WANG W H， ZOU S Y，et al ．Joint noise reduction and feature extraction of acoustic emission signals for hydraulic turbines under cavitation[J]．Journal of Hydroelectric Engineering，2022，41(12)：145-152．
[5]	ZHU Y， ZHOU L， LV S，et al ．Research progress on identification and suppression methods for monitoring the cavitation state of centrifugal pumps[J]．Water，2023，16(1)：52． doi:10.3390/w16010052
[6]	BUI T T， VU V H， LIU Z，et al ．Updating singular value decomposition for modal analysis of slow-varying non-stationary vibration structures[J]．Journal of Vibration and Control，2023：308． doi:10.1177/10775463231214308
[7]	蒋文君，胡晓，张培，等．基于EEMD近似熵的水电机组振动信号特征提取[J]．水力发电学报，2020，39(6)：18-27．
	JIANG W J， HU X， ZHANG P，et al ．Vibration feature extraction for hydropower units based on ensemble empirical mode decomposition and approximate entropy[J]．Journal of Hydroelectric Engineering，2020，39(6)：18-27．
[8]	BAI S， ZENG Y，DAO F，et al ．Signal spectrum analysis of sediment water impact of hydraulic turbine based on ICEEMDAN-wavelet threshold denoising strategy[J]．Water，2023，15(22)：4017． doi:10.3390/w15224017
[9]	邵必林，纪丹阳．基于VMD-SE的电力负荷分量的多特征短期预测[J]．中国电力，2024，57(4)：162-170．
	SHAO B L， JI D Y ．Multi-feature short-term prediction of power load components based on VMD-SE[J]．Electric Power，2024，57(4)：162-170．
[10]	LIU D， CUI L， WANG H ．Rotating machinery fault diagnosis under time-varying speeds：a review[J]．IEEE Sensors Journal，2023，23(24)：29969-29990． doi:10.1109/jsen.2023.3326112
[11]	李楠，姜涛，隋想，等．一种时频尺度下的多元短期电力负荷组合预测方法[J]．电力系统保护与控制，2024，52(13)：47-58．
	LI N， JIANG T， SUI X，et al ．A multi-component short-term power load combination forecasting method on a time-frequency scale[J]．Power System Protection and Control，2024，52(13)：47-58．
[12]	陈强，李茂峰，秦际明，等．基于改进残差网络的XLPE电缆局部放电声纹诊断方法[J]．广东电力，2024，37(5)：97-103．
	CHEN Q， LI M F， QIN J M，et al ．Diagnosis method of voiceprint of partial discharge in XLPE cable based on improved ResNet[J]．Guangdong Electric Power，2024，37(5)：97-103．
[13]	刘展，包琰洋，李大字．基于重采样降噪与主成分分析的宽卷积深度神经网络风机故障诊断方法[J]．发电技术，2023，44(6)：824-832．
	LIU Z， BAO Y Y， LI D Z ．Fault diagnosis method of wind turbines based on wide deep convolutional neural network with resampling and principal component analysis[J]．Power Generation Technology，2023，44(6)：824-832．
[14]	孙文成，李健，彭宇辉，等．基于样本不均衡和特征优选多源融合的输电线路故障类型辨识[J]．电测与仪表，2024，61(12)：79-89．
	SUN W C， LI J， PENG Y H，et al ．Transmission line fault type identification based on the sample imbalance and feature preferred multi-source fusion[J]．Electrical Measurement ＆ Instrumentation，2024，61(12)：79-89．
[15]	罗国敏，杨雪凤，尚博阳，等．基于改进堆叠降噪自编码器的配电网高阻接地故障检测方法[J]．电力系统保护与控制，2024，52(24)：149-160．
	LUO G M， YANG X F， SHANG B Y，et al ．High impedance grounding fault detection method of a distribution network based on an improved stacked denoised autoencoder[J]．Power System Protection and Control，2024，52(24)：149-160．
[16]	李兵，梁舒奇，单万宁，等．基于改进正余弦算法优化堆叠降噪自动编码器的电机轴承故障诊断[J]．电工技术学报，2022，37(16)：4084-4093．
	LI B， LIANG S Q， SHAN W N，et al ．Motor bearing fault diagnosis based on improved sine and cosine algorithm for stacked denoising autoencoders[J]．Transactions of China Electrotechnical Society，2022，37(16)：4084-4093．
[17]	黄力，宋爽，刘闯，等．基于IHHO-HKELM输电线路覆冰预测模型[J]．电力科学与技术学报，2024，39(4)：33-41．
	HUANG L， SONG S， LIU C，et al ．Icing prediction model of transmission line based on IHHO-HKELM[J]．Journal of Electric Power Science and Technology，2024，39(4)：33-41．
[18]	邱书琦，蹇照民，方立雄，等．基于变分模态分解和集成学习的光伏发电预测[J]．智慧电力，2024，52(3)：32-38．
	QIU S Q， JIAN Z M， FANG L X，et al ．Photovoltaic power generation forecasting based on variational modal decomposition and ensemble learning[J]．Smart Power，2024，52(3)：32-38．
[19]	AZAMI H， ESCUDERO J ．Amplitude-and fluctuation-based dispersion entropy[J]．Entropy，2018，20(3)：210． doi:10.3390/e20030210
[20]	杨康，李蓝青，李艺丰，等．一种新型分布式光伏出力区间预测方法[J]．发电技术，2024，45(4)：684-695．
	YANG K， LI L Q， LI Y F，et al ．A novel distributed photovoltaic output interval prediction method[J]．Power Generation Technology，2024，45(4)：684-
[695]	. doi:10.1023/a:1010650624155
[21]	赵妍，潘怡，李亚波，等．基于AVMD多尺度模糊熵和VPMCD算法的宽频振荡分类[J]．电力系统保护与控制，2024，52(13)：179-187．
	ZHAO Y， PAN Y， LI Y B，et al ．Broadband oscillation classification based on AVMD multi-scale fuzzy entropy and the VPMCD algorithm[J]．Power System Protection and Control，2024，52(13)：179-187．
[22]	王子江，张兆毅，樊友平，等．基于近似熵变化量判据的混合直流输电系统纵联保护方案[J]．电力系统保护与控制，2024，52(21)：1-12．
	WANG Z J， ZHANG Z Y， FAN Y P，et al ．A pilot protection scheme for a hybrid HVDC transmission system based on an approximate entropy change criterion[J]．Power System Protection and Control，2024，52(21)：1-12．
[23]	LAROCHELLE H， BENGIO Y， LOURADOUR J，et al ．Exploring strategies for training deep neural networks[J]．Journal of Machine Learning Research，2009，10：1-40．
[24]	刘忠，邹淑云，陈莹，等．混流式水轮机模型空化状态与声发射信号特征关系试验[J]．动力工程学报，2016，36(12)：1017-1022．
	LIU Z， ZOU S Y， CHEN Y，et al ．Experiments on the relationship between cavitation status and acoustic emission signal features for a francis turbine model[J]．Journal of Chinese Society of Power Engineering，2016，36(12)：1017-1022．