Cavitation State Recognition of Regulating Valve Based on Improved SVM Algorithm

doi:10.3901/JME.2023.24.312

Abstract

Abstract: The cavitation often causes vibration, noise and reduction of working efficiency of the regulating valve, which will seriously affect the safe operation and component life of the coal liquefaction system. Accurately identifying the cavitation state in the regulating valve can provide data support for monitoring the cavitation state in the regulating valve and predictive maintenance of the regulating valve. Due to it is difficult to effectively identify the cavitation state of the regulating valve, a support vector machines(Support vector machines, SVM) model based on genetic algorithm and kernel principle component analysis(Kernel principle component analysis, KPCA) is proposed to identify the cavitation state of the regulating valve in this paper. The characteristics of the vibration signal are extracted by time domain, frequency domain and wavelet packet transform, the principal components of the feature vector are extracted, then the SVM optimized by genetic algorithm is used for cavitation state Recognition of regulating valve. The experimental results show that KPCA can effectively extract the nonlinear principal components of the eigenvectors of vibration signals, and the constructed SVM can effectively identify the cavitation state of the regulating valve. Compared with the neural network, the improved SVM has better recognition effect, whose identification accuracy rate could reach to 98.7%.

Key words: regulating valve, cavitation status recognition, SVM, genetic algorithm

CLC Number:

TH137

LI Beibei, SUN Shenzhen, LIU Xiumei, LIU Qihang, LIU Shen, ZHAO Qiao, HE Jie. Cavitation State Recognition of Regulating Valve Based on Improved SVM Algorithm[J]. Journal of Mechanical Engineering, 2023, 59(24): 312-322.

References

[1] 张微,许学超,余道刚. 静流元件调节阀气动及振动噪声性能试验研究[J]. 热能动力工程,2019,34(1):82-86. ZHANG Wei,XU Xuechao,YU Daogang. Experimental study on aerodynamic and vibration noise performance of quiet flow element valve[J]. Journal of Engineering for Thermal Energy and Power,2019,34(1):82-86.
[2] SHA Y,FABER J,GOU S P,et al. A multi-task learning for cavitation detection and cavitation intensity recognition of valve acoustic signals[J]. Engineering Applications of Artificial Intelligence,2022,113:104904.
[3] HABIBNEJAD D,AKBARZADEH P,SALAVATIPOUR A,et al. Cavitation reduction in the globe valve using oblique perforated cages:A numerical investigation[J]. Flow Measurement and Instrumentation,2022,83:102110.
[4] 闵为,王东,郑直,等. 低压下锥阀振荡空化的可视化试验研究[J]. 机械工程学报,2018,54(20):139-144. MIN Wei,WANG Dong,ZHENG Zhi,et al. Visualization experiment of vibration and cavitation in poppet valve under low pressure condition[J]. Journal of Mechanical Engineering,2018,54(20):139-144.
[5] 明廷锋,曹玉良,贺国,等. 流体机械空化检测研究进展[J]. 武汉理工大学学报(交通科学与工程版),2016,40(2):219-226. MING Tingfeng,CAO Yuliang,HE Guo,et al. Research progress of fluid machinery cavitation detection[J]. Journal of Wuhan University of Technology(Transportation Science & Engineering),2016,40(2):219-226.
[6] CUDINA M. Detection of cavitation phenomenon in a centrifugal pump using audible sound[J]. Mechanical Systems and Signal Processing,2003,17(6):1335-1347.
[7] MCKEE K K,FORBES G L,MAZHAR I,et al. A vibration cavitation sensitivity parameter based on spectral and statistical methods[J]. Expert Systems with Applications,2015,42(1):67-78.
[8] 段向阳,王永生,苏永生. 振动分析在离心泵空化监测中的应用[J]. 振动与冲击,2011,30(4):161-165. DUAN Xiangyang,WANG Yongsheng,SU Yongsheng. Vibration analysis applied in cavitation monitoring of a centrifugal pump[J]. Journal of Vibration and Shock,2011,30(4):161-165.
[9] 苏永生,王永生,段向阳. 基于多传感器数据融合的喷水推进泵空化分类识别[J]. 振动与冲击,2012,31(18):93-95,121. SU Yongsheng,WANG Yongsheng,DUAN Xiangyang. Classification of cavitation in a waterjet pump based on multi-sensor data fusion[J]. Journal of Vibration and Shock,2012,31(18):93-95,121.
[10] AZIZI R,ATTARAN B,HAJNAYEB A,et al. Improving accuracy of cavitation severity detection in centrifugal pumps using a hybrid feature selection technique[J]. Measurement,2017,108:9-17.
[11] PANDA A K,RAPUR J S,TIWARI R. Prediction of flow blockages and impending cavitation in centrifugal pumps using support vector machine(SVM) algorithms based on vibration measurements[J]. Measurement,2018,130:44-56.
[12] YAN Z L,LIU J,CHEN B,et al. Fluid cavitation detection method with phase demodulation of ultrasonic signal[J]. Applied Acoustics,2015,87:198-204.
[13] 曹玉良,明廷锋,贺国,等. 基于深度学习的离心泵空化状态识别[J]. 西安交通大学学报,2017,51(11):165-172. CAO Yuliang,MING Tingfeng,HE Guo,et al. Cavitation state recognition of centrifugal pump based on deep learning[J]. Journal of Xi'an Jiaotong University,2017,51(11):165-172.
[14] LI B,LI R,YAN J,et al. Vibration characteristics of the two-stage throttle valve induced by cavitation[J]. The Journal of Engineering,2019,2019(13):102-106.
[15] MA H,ZHANG T,AN Q,et al. Visualization experiment and numerical analysis of cavitation flow characteristics in diesel fuel injector control valve with different structure design[J]. Journal of Thermal Science,2021,30(1):76-78.
[16] SHI J,REN Y,TANG H,et al. Hydraulic directional valve fault diagnosis using a weighted adaptive fusion of multi-dimensional features of a multi-sensor[J]. Journal of Zhejiang University-Science A(Applied Physics & Engineering),2022,23(4):257-272.
[17] 谭树彬,刘建昌,钟云峰,等. 基于神经网络的电液伺服阀故障诊断[J]. 仪器仪表学报,2006(S1):401-403,418. TAN Shubin,LIU Jianchang,ZHONG Yunfeng,et al. Fault di agnosis of electro-hydraulic servo val ve based on neural net work[J]. Chinese Journal of Scientific Instrument,2006(S1):401-403,418.
[18] 李瑞,范玉刚. 基于CEEMDAN多尺度排列熵和SO-RELM的高压隔膜泵单向阀故障诊断[J]. 振动与冲击,2023,42(5):127-135. LI Rui,FAN Yugang. Fault diagnosis of one-way valve of high-pressure diaphragm pump based on CEEMDAN multi-scale permutation entropy and SO-RELM[J]. Journal of Vibration and Shock,2023,42(5):127-135.
[19] 陆亮. 液压节流阀中的空化流动与噪声[D]. 杭州:浙江大学,2012. LU Liang. Cavitaion flow and noise in hydraulic throttling valves[D]. Hangzhou:Zhejiang University,2012.
[20] 孙原理,宋志浩. 基于多物理场信号相关分析与支持向量机的离心泵故障诊断方法[J]. 振动与冲击,2022,41(6):206-212. SUN Yuanli,SONG Zhihao. Centrifugal pump fault diagnosis based on the multiple-physicail fieldsignal correalation analysis and suppot vector machine[J]. Journal of Vibration and Shock,2022,41(6):206-212.
[21] 胡晓强,仲训昱,张霄力,等. 基于支持向量机辅助的四轴陀螺两级故障诊断方法[J]. 上海交通大学学报,2020,54(11):1151-1156. HU Xiaoqiang,ZHONG Xunyu,ZHANG Xiaoli,et al. A two-level fault diagnosis method for gyro-quadruplet assisted by support vector machine[J]. Journal of Shanghai Jiao Tong University,2020,54(11):1151-1156.
[22] 李向杰,张向文. 基于人工蜂群支持向量机的电动汽车制动意图识别方法[J]. 中国机械工程,2021,32(17):2125-2135. LI Xiangjie,ZHANG Xiangwen. Recognition method of branking intention of electric vehicles based on ABC-SVM algorithm[J]. China Mechanical Engineering,2021,32(17):2125-2135.