水轮机压力脉动是水电机组运行过程中不可避免的现象,准确地识别和定量诊断脉动状态对机组高效稳定运行尤为重要。为此,本文提出了基于水电机组运行工况的水轮机压力脉动诊断策略,以水电机组实际运行工况为切入点,通过分析工况参数与压力脉动的非线性相关关系,得到影响压力脉动的主要相关工况参数,提取了融合机组运行工况参数与脉动幅值特性的特征向量,并利用支持向量机(SVM)与极限学习机(ELM)两种诊断方法进行脉动状态定性诊断。研究压力脉动幅值历史统计规律,提出了脉动状态对机组劣化程度的模糊评估函数,反演了定性诊断结果与机组健康状态的映射关系,实现压力脉动的定量诊断。实例验证表明,相对于仅基于脉动幅值的诊断策略而言,该方法诊断准确率更高,定量诊断指标可靠有效。这为水电机组安全稳定运行提供技术保障。
Abstract
Pressure pulsation during the operation of hydroelectric generating unit (HGU) is inevitable phenomenon. Diagnosing and assessing accurately pulsation state are of particular importance. In view of the pressure pulsation is closely related to the operating state of HGU, a novel diagnosis strategy based on working condition is proposed in this paper: Firstly, contribution rates based on mutual information analysis is computed to extract the superior condition parameters, and this superior condition parameters and time-frequency of pulsation signals are fused, the fusion information are considered to the eigenvectors of pressure pulsation. Then, support vector machine (SVM) and extreme learning machine (ELM) are been used to diagnose the pulsation state. Finally, in order to achieve a quantitative diagnosis for pressure pulsation, the degradation function is given with fuzzy evaluation theory. The results of a real example show that this diagnosis strategy is better than the traditional time-frequency diagnosis strategy, and it is of practical guiding significance to safety and stable operation of unit to quantify pulsation assessment.
关键词
水电机组 /
水轮机 /
压力脉动 /
故障诊断 /
定量诊断 /
支持向量机(SVM) /
极限学习机(ELM)
{{custom_keyword}} /
Key words
hydroelectric generating unit /
hydraulic turbine /
pressure pulsation /
fault diagnosis /
quantitative diagnosis /
support vector machine (SVM) /
extreme learning machine (ELM)
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 田锋社. 水轮机压力脉动测试的分析与探讨 [J]. 水利水电科技进展, 2006, 26(2): 37-39.
Feng-she T.Study on pressure fluctuation test of water turbine[J].Advances in Science and Technology of Water Resources,2006, 26(2): 37-39.
[2] 张瑞金, 徐洪泉, 张建光, 等. 水轮机压力脉动的测试分析与评价[J]. 水利水电技术, 2003, 34(11): 76-78.
Zhang Ruijin,Xu Hongquan,Zhang Jianguang,et al.Analyzing and Evaluating on Pressure Fluctuation Measurement of Hydro Turbines[J].Water Resources and Hydropower Engineering,2003, 34(11):76-78.
[3] 冯志鹏, 褚福磊. 基于 Hilbert-Huang变换的水轮机非平稳压力脉动信号分析[J]. 中国电机工程学报, 2005, 25(10): 111-115.
FENG Z,CHU F.TRANSIENT HYDRAULIC PRESSURE FLUCTUATION SIGNAL ANALYSIS OF HYDROTURBINE BASED ON HILBERT-HUANG TRANSFORM [J].Proceedings of the Csee,2005,25(10):111-115.
[4] 严雄. Laplace 相关滤波法在水电机组状态分析中的应用研究[D]. 西安:西安理工大学, 2007.
YAN Xiong.APPLICATION RESEARCH OF LAPLACE WAVELET CORRELATION FILTERING METHOD INTO STATE ANALYSIS OF HYOROELCTRIC SET[J].Xi'an:Xi'an University of Technology,2007 (in Chinese).
[5] 赵林明, 楚清河, 代秋平, 等. 基于小波分析与人工神经网络的水轮机压力脉动信号分析[J]. 水利学报, 2011, 42(9): 1075-1080.
Zhao L M,Chu Q H,Dai Q P, et al.Analysis of pressure fluctuation in draft tube based on wavelet analysis and artificial neural networks[J].Journal of Hydraulic Engineering,2011,42(9):1075-1080.
[6] 王瀚, 张欣伟, 罗兴锜, 等. EMD 动态过程神经网络尾水管故障信息融合诊断[J]. 水力发电学报, 2012, 31(006): 282-287.
Wang H,Zhang X,Luo X,et al.Process neural network based on EMD for fault fusion diagnosis of draft tube[J].Shuili Fadian Xuebao(Journal of Hydroelectric Engineering) ,2012,31(006):282-287.
[7] 程建, 李友平, 彭兵, 等. 水电机组状态检修发展思路探讨[J]. 西北水电, 2012, 1.
CHENG Jian,LI Youping,PENG Bin,et al.Study on Development Concepts of Turbine-Generator State Maintenance[J].XIBEI SHUI DIAN,2012,1.
[8] Heng A, Zhang S, Tan A C C, et al. Rotating machinery prognostics: State of the art, challenges and opportunities[J]. Mechanical Systems and Signal Processing, 2009, 23(3): 724-739.
[9] 杨智春, 陈帅,李斌. 基于统计与粗糙集理论的飞机垂尾抖振载荷分布假设选择与评价方法[J]. 振动与冲击, 2012,31 (5): 6-11.
YANG Zhi-chun,CHEN Shuai,LI Bin.Selection and evalution method for distribtuion assumptions of aircraft buffet load based on statistics and rough set theory[J].JOURNAL OF VIBRATION AND SHOCK,2012,31 (5):6-11.
[10] 白 斌,白广忱,林学柱.基于 FSVM 改良隶属度的发动机振动故障识别[J] . 振动与冲击, 2013, 32(20): 23-28.
BAI Bin,BAI Guang-chen,LIN Xue-zhu.Improved FSVM and multi-class fuzzy membership method for aeroengine vibration fault identification[J].JOURNAL OF VIBRATION AND SHOCK,2013,32(20):23-28.
[11] 赵志宏, 杨绍普, 申永军. 基于独立分量分析与相关系数的机械故障特征提取[J]. 振动与冲击, 2013, 32(6): 67-72.
ZHAO Zhi-hong,YANG Shao-pu,SHEN Yong-jun.Machinery fault feature extraction based on independent component analysis and correlation coefficient[J].JOURNAL OF VIBRATION AND SHOCK,2013,32(6):67-72.
[12] Li N, Zhou R, Hu Q, et al. Mechanical fault diagnosis based on redundant second generation wavelet packet transform neighborhood rough set and support vector machine[J]. Mechanical Systems and Signal Processing, 2012,28: 608-621.
[13] 高光勇, 蒋国平. 采用优化极限学习机的多变量混沌时间序列预测[J]. 物理学报, 2012, 61(4): 37-45.
GAO Guang-yong,JIANG Guo-ping.Prediction of multivariable chaotic time series using optimized extreme learning machine [J].Acta Physica Sinica,2012,64(4):37-54.
[14] Huang G B, Zhu Q Y, Siew C K. Extreme learning machine: theory and applications[J]. Neurocomputing, 2006, 70(1): 489-501.
[15] 赵林明, 魏德华, 何成连. 水轮机尾水管压力脉动的神经网络模型[J]. 水利学报, 2005, 36(11): 1375-1378.
Zhao L M,Chu Q H,Dai Q P, et al.Neural network model for pressure fluctuation in draft tube of water turbine[J].Journal of Hydraulic Engineering,2005,36(11):1375-1378.
[16] 胡永祥. 基于互信息的多模态医学图像非刚性配准研究 [D]. 中南大学, 2007.
HU Yangxiang. Research on multimodality medical image non-rigid registration based on mutual information[D].Changsha:Central South University,2007 (in Chinese).
[17] Kai-Yuan, C., Chuan-Yuan, W. & Ming-Lian Z. (1991). Posbist reliability behavior of typical systems with two types of failure[J]. Fuzzy sets and systems, 43(1), 17-32.
[18] 殷世茂. 系统模糊可靠性[D]. 合肥: 安徽工业大学, 2012.
YIN Shimao. Fuzzy reliability of system[D].Hefei:Anhui University of Technology,2012 (in Chinese).
[19] 顾煜炯, 董玉亮, 杨昆. 基于模糊评判和 RCM 分析的发电设备状态综合评价[J]. 中国电机工程学报, 2004, 24(6): 189-194.
GU Y,DONG Y,Yang K.Synthetic evaluation on conditions of equipment in power plant based on Fuzzy judgment and RCM analysis [J].Proceedings of the Csee,2004,24(6):189-194.
[20] 罗华军, 刘德富, 黄应平.基于遗传算法-支持向量机的水库叶绿素a浓度短期预测非线性时序模型[J]. 水利学报, 2009, 40(1): 46-51.
LUO Huajun,LIU Defu, HUANG Yingping.Genetic algorithm-support vector machine model for short-term prediction of chlorophyll a concentration nonlinear time series [J].JOURNAL OF HYDRAULIC ENGINEERING,2009,40(1):46-51.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}