阀门内漏单声波传感器监测方法研究

doi:10.11949/j.issn.0438-1157.20190007

摘要/Abstract

摘要：

针对阀门内漏模拟和阀门内漏监测问题，提出了阀门内漏模拟和阀门内漏非介入式单声波传感器监测方法。设计了由法兰、薄纸片和带孔铝片组成的虚拟阀门，可真实地模拟阀门内漏瞬态过程，克服了传统阀门内漏模拟方法中存在的各种缺点。分析、比较了不同条件下阀门上下游声波传感器输出信号的频域特征，验证了阀门上游声波传感器设置的冗余性和基于单声波传感器实现阀门内漏监测的可行性。分析、比较了不同条件下单声波传感器输出信号的时频域特征，提出了基于峰值系数和小波包能量分率的阀门内漏时频域特征提取方法，并以阀门密封良好状态下的声波信号为样本，建立了基于SVDD的阀门内漏诊断模型。测试结果表明：本文提出的单声波传感器阀门内漏监测方法可行、模型诊断准确，且具有较强的抗干扰能力和较高的阀门内漏检测灵敏度。

关键词: 阀门内漏, 非介入式, 单传感器, 声波监测, 虚拟阀门

Abstract:

Aiming at simulation and monitoring for valve internal leakage, a single-senor based non-intrusive acoustic monitoring method for valve internal leakage and a valve internal leakage simulation method are proposed. A virtual valve composed of a flange, thin paper sheet and perforated aluminum sheet is designed, which can truly simulate the transient leakage of a valve and overcome shortcomings in traditional valve internal leakage simulation method. By analyzing and comparing the frequency domain characteristics of the output signal of the upstream and downstream acoustic sensors under different conditions, the redundancy of the upstream acoustic sensor and the feasibility of single-sensor based acoustic monitoring method for valve internal leakage are verified. The time-frequency domain characteristics of the output signal of single acoustic sensor under different conditions are analyzed and compared, a time-frequency domain feature extraction method for acoustic signals based on wavelet packet energy fractionation and the kurtosis coefficient is proposed, the valve internal leakage diagnosis model based on support vector data description method (SVDD) was established with the normal samples under valve closed condition. The experimental results show that the proposed method for valve internal leakage monitoring with single acoustic sensor is effective, and it has strong anti-interference ability and high detection sensitivity.

Key words: valve internal leakage, non-intrusive, single-sensor structure, acoustic monitoring, virtual valve

中图分类号:

TE 988

王文亮, 王芳, 吴石恩, 林伟国. 阀门内漏单声波传感器监测方法研究[J]. 化工学报, 2019, 70(S1): 248-256.

Wenliang WANG, Fang WANG, Shi en WU, Weiguo LIN. Research on valve internal leakage monitoring based on single acoustic sensor[J]. CIESC Journal, 2019, 70(S1): 248-256.

图/表 19

图1 阀门内漏监测系统实验平台

Fig.1 Experimental platform of valve internal leakage monitoring system

图2 虚拟阀门内部结构

Fig.2 Internal structure of virtual valve

图3 阀门密封良好时上、下游声波信号频域特征对比

Fig.3 Comparison of frequency domain characteristics of upstream and downstream acoustic signals under valve normal condition

图4 阀门内漏时上、下游声波信号频域特征对比

Fig.4 Comparison of frequency domain characteristics of upstream and downstream acoustic signals under valve internal leakage

图5 启泵时上、下游声波信号频域特征对比

Fig.5 Comparison of frequency domain characteristics of upstream and downstream acoustic signals under pump starting condition

图6 阀门密封良好时与阀门内漏发生时下游声波信号

Fig.6 Downstream acoustic signals under valve normal and internal leakage conditions

表1 实验条件

Table 1 Test condition

参数名称	数值
阀门上游压力	200 kPa
阀门下游压力	101 kPa
阀门泄漏孔径	1.7%阀门开度
管道直径	DN20
管壁厚度	6 mm
信号采样频率fs	100 Hz

图7 阀门密封良好与阀门内漏发生时下游声波信号概率密度曲线

Fig.7 Probability density curves of downstream acoustic signals under valve normal and internal leakage conditions

表2 声波信号时域特征参数

Table 2 Acoustic signal time domain characteristic parameters

序号	均方根值（RMS）/mV	峰度系数（K_u）
1	36.3598	3.3304
2	37.0721	3.4176
3	36.4608	3.3501
4	36.8929	3.3946
5	36.3821	3.2689
6	124.4467	7.1149
7	118.4837	11.1468
8	125.7384	9.5663
9	116.4211	8.3873
10	131.8745	9.9753

图8 不同管道压力下声波信号的均方根值及峰度系数

Fig.8 Values of RMS and kurtosis coefficient of acoustic signal under different pipeline pressure conditions

图9 阀门密封良好和内漏状态下声波信号功率谱图

Fig.9 Power spectrum of acoustic signal under valve normal and internal leakage conditions

表3 阀门密封良好和内漏状态下信号的功率谱重心

Table 3 Power spectrum gravity under valve normal and internal leakage conditions/Hz

序号	阀门正常	瞬态内漏
1	16.4332	5.7205
2	16.7183	8.5889
3	16.7863	3.9855
4	16.6846	5.8735
5	16.454	6.7332
6	16.7299	4.1248
7	17.3117	6.4734
8	16.4612	4.2227
9	17.479	7.0074
10	17.2883	4.2243
均值	16.8347	5.6954

图10 阀门密封良好和发生内漏时声波信号的小波包能量分率

Fig.10 Wavelet packet energy rate of acoustic signal under valve normal and internal leakage conditions

图11 第一类误差和超球体半径随核参数σ的变化

Fig.11 First type of error and radius of hypersphere vary with kernel parameter σ

表4 测试样本集

Table 4 Test sample set

管道压力/kPa	正常样本帧数	内漏样本帧数
100	90	5
200	70	10
300	85	5

图12 SVDD测试结果

Fig.12 Results of SVDD model test

图13 启停泵测试

Fig.13 Start-stop pump test

图14 环境噪声测试结果

Fig.14 Results of environmental noise test

表5 检测灵敏度测试实验结果

Table 5 Test results of sensitivity test

泄漏孔径/mm	阀门上游压力/kPa	平均泄漏量/（L/min）	模型输出f(z) （第1/2次）
0.8	210	0.328	0.0234/0.0216
1	240	0.540	0.0426/0.0666
2	220	1.933	0.8905/0.6025
3	200	4.375	0.5896/0.9449

参考文献 18

1	王忠锋, 黄剑龙, 李力刚, 等. 基于无线通信的阀门泄漏检测系统[J]. 仪器仪表学报, 2009, 30(6): 517-519.
	WangZ F, HuangJ L, LiL G, et al. Valve leakage detection system based on wireless communication [J]. Chinese Journal of Scientific Instrument, 2009, 30(6): 517-519.
2	LeeJ H, LeeM R, KimJ T, et al. A study of the characteristics of the acoustic emission signals for condition monitoring of check valves in nuclear power plants [J]. Nuclear Engineering and Design, 2006, 236(13): 1411-1421.
3	LeeM R, LeeJ H, KimJ T. Condition monitoring of a nuclear power plant check valve based on acoustic emission and a neural network [J]. Journal of Pressure Vessel Technology, 2005, 127(3): 230-236.
4	LeeS G, ParkJ H, YooK B, et al. Evaluation of internal leak in valve using acoustic emission method [J]. Key Engineering Materials, 2006, 326/327/328: 661-664.
5	KaewwaewnoiW, PrateepasenA , KaewtrakulpongP. Investigation of the relationship between internal fluid leakage through a valve and the acoustic emission generated from the leakage[J]. Measurement, 2010, 43(2): 274-282.
6	李振林, 张海峰, 夏广辉. 基于声发射理论的阀门气体内漏量化检测研究[J]. 振动与冲击, 2013, 32(15): 77-81.
	LiZ L, ZhangH F, XiaG H. Quantitative detection of valve internal air leakage based on acoustic emission theory [J]. Journal of Vibration and Shock, 2013, 32(15): 77-81.
7	ThompsonG, ZolkiewskiG. An experimental investigation into the detection of internal leakage of gases through valves by vibration analysis [J]. Journal of Hubei University of Economics, 2004, 211(3): 195-207.
8	YerraM, NasipuriA, DuemmlerH. Sequential state logic for pneumatic valve monitoring using piezo film sensors[C]// SoutheastCon. IEEE, 2017: 1-8.
9	林伟国, 吴石恩. 阀门内漏声波监测及其信号特征提取方法[J]. 振动与冲击, 2018, 37(9): 120-126.
	LinW G, WuS E. Acoustic monitor and signal feature extraction for valve internal leakage [J]. Journal of Vibration and Shock, 2018, 37(9): 120-126.
10	王超, 凌志浩, 林琦彬. 无线阀门泄漏变送器的设计[J]. 仪器仪表学报, 2010, 31(8): 75-79.
	WangC, LingZ H, LinQ B. Design of wireless valve leakage transmitter[J].Chinese Journal of Scientific Instrument, 2010, 31(8): 75-79.
11	刘贵杰, 徐萌, 王欣, 等. 基于HHT的管道阀门内漏声发射检测研究[J]. 振动与冲击, 2012, 31(23): 62-66.
	LiuG J, XuM, WangX, et al. AE detection for pipeline valve leakage based on HHT [J]. Journal of Vibration and Shock, 2012, 31(23): 62-66.
12	宗福兴, 税爱社, 汪辉, 等. 基于CCA和WT的油库阀门内漏声发射信号去噪 [J]. 仪器仪表学报, 2014, 35(9): 2004-2011.
	ZongF X, ShuiA S, WangH, et al. Denoising method for acoustic emission signal in oil depot internal valve leakage inspection based on CCA and WT [J]. Chinese Journal of Scientific Instrument, 2014, 35(9): 2004-2011.
13	ZhangH F, LiZ L, JiZ L, et al. Intelligent leak level recognition of gas pipeline valve using wavelet packet energy and support vector machine model [J]. Insight: Non-Destructive Testing and Condition Monitoring, 2013, 55(12): 670-674.
14	BalandaK, MacgillivrayH L. Kurtosis: a critical review [J]. American Statistician, 1988, 42(2): 111-119.
15	GuoM, LinW G, WuH Y, et al. An early-warning method for agglomeration detection in gas-solid fluidized bed based on hydromechanics [J]. Applied Mechanics & Materials, 2013, 387: 152-158.
16	TaxD M J, DuinR P W. Support Vector Data Description [M]. Kluwer Academic Publishers, 2004.
17	SmolaA J, SchölkopfB, MüllerK R. The connection between regularization operators and support vector kernels [J]. Neural Network, 1998, 11(4): 637-649.
18	毛荣富, 朱海潮, 何琳, 等. 一种基于SVDD的舰船机械特征选取新方法[J]. 振动与冲击, 2008, 27(4): 101-104.
	MaoR F, ZhuH C, HeL, et al. A new feature selection method for machines on shipboard based on support vector data description [J]. Journal of Vibration and Shock, 2008, 27(4): 101-104.