基于多参数融合的PSO-GBDT空化状态识别方法研究

吴思宇; 李贝贝; 刘秀梅; 韩蕊; 徐爽; 王蕊

doi:null

您当前的位置：

首页 >

文章列表页 >

基于多参数融合的PSO-GBDT空化状态识别方法研究

更新时间：2026-09-30

- 基于多参数融合的PSO-GBDT空化状态识别方法研究
- Research on Cavitation State Recognition Method Based on Multi-Parameter Fusion Using PSO-GBDT
- 西安交通大学学报 2026年
- 作者机构：
  
  中国矿业大学机电工程学院，江苏，徐州 221116
- 作者简介：
- 基金信息：
- DOI：
  中图分类号：
- 收稿：2026-07-13，
  
  修回：2026-09-08，
  
  录用：2026-09-14，
- 稿件说明：
移动端阅览
吴思宇, 李贝贝, 刘秀梅, 等. 基于多参数融合的PSO-GBDT空化状态识别方法研究[J/OL]. 西安交通大学学报,2026-09-30. http://zkxb.xjtu.edu.cn/zh/article/173030479/.

WU Siyu, LI Beibei, LIU Xiumei, et al. Research on Cavitation State Recognition Method Based on Multi-Parameter Fusion Using PSO-GBDT[J/OL]. JOURNAL OF XI’AN JIAOTONG UNIVERSITY,2026-09-30. http://zkxb.xjtu.edu.cn/zh/article/173030479/.
吴思宇, 李贝贝, 刘秀梅, 等. 基于多参数融合的PSO-GBDT空化状态识别方法研究[J/OL]. 西安交通大学学报,2026-09-30. http://zkxb.xjtu.edu.cn/zh/article/173030479/. DOI：

WU Siyu, LI Beibei, LIU Xiumei, et al. Research on Cavitation State Recognition Method Based on Multi-Parameter Fusion Using PSO-GBDT[J/OL]. JOURNAL OF XI’AN JIAOTONG UNIVERSITY,2026-09-30. http://zkxb.xjtu.edu.cn/zh/article/173030479/. DOI：

摘要

煤液化系统中，调节阀在大压差节流工况下易诱发空化，导致通流能力衰减与压力脉动增强，准确识别调节阀的空化状态，对于防范空化引发的通流能力衰减、抑制压力脉动，以及保障煤液化装置的长周期稳定运行具有重要意义。为实现空化状态的高效识别，本文搭建了调节阀空化流动实验平台，并选取开度、进出口压差、进出口压力值、流量、流量系数、偏差比及压力脉动均方根值（RMS）等流体特征作为输入变量，构建基于多参数融合的空化状态识别模型。对比评估随机森林（RF）、梯度提升决策树（GBDT）与极限梯度提升（XGBoost）后发现，GBDT综合性能最优，准确率达90.61%。进一步引入粒子群优化算法（PSO）建立PSO-GBDT 模型，将准确率提升至92.12%，降低了初生空化与临界空化之间的误判，且其性能显著优于任意单一特征的识别准确率。基于重要性评估结果表明，偏差比与进出口压差对分类贡献最大，在剔除开度、进口压力等低贡献度冗余特征后，模型准确率仍维持在91.52%的高水平。研究表明，所构建的多参数融合模型能够实现调节阀空化状态的有效判别，为系统的在线监测提供方法支撑。

Abstract

In coal liquefaction systems

regulating valves are prone to cavitation under large pressure-drop throttling conditions

leading to degraded flow capacity and intensified pressure pulsations

which ultimately undermine the long-term stable operation of the plant. To achieve the efficient identification of cavitation stages

an experimental platform for cavitating flow in a regulating valve was established. Fluid features—including valve opening

pressure drop

inlet and outlet pressures

flow rate

flow coefficient

deviation ratio

and the root-mean-square (RMS) value of pressure pulsations—were selected as input variables to construct a cavitation stage identification model based on multi-parameter fusion. A comparative evaluation of Random Forest (RF)

Gradient Boosting Decision Tree (GBDT)

and Xtreme Gradient Boosting (XGBoost) reveals that GBDT exhibits the best overall performance with an accuracy of 90.61%. Furthermore

the Particle Swarm Optimization (PSO) algorithm was introduced to develop a PSO-GBDT model

which improves the accuracy to 92.12% and reduces the misclassification between incipient cavitation and critical cavitation. Moreover

its classification performance significantly outperforms the accuracy of any single-feature model. Feature importance analysis indicates that the deviation ratio and pressure drop contribute the most to the classification. Based on the importance evaluation

after eliminating low-contribution redundant features such as valve opening and inlet/outlet pressures

the model accuracy is still maintained at a high level of 91.52%. The study demonstrates that the proposed multi-parameter fusion model enables the effective identification of cavitation stages in regulating valves

providing methodological support for the online monitoring of the system.

关键词

Keywords

references

HE J , ZHANG Y , LIU X , et al . Experiment and simulation study on cavitation flow in pressure relief valve at different hydraulic oil temperatures [J ] . Flow Measurement and Instrumentation , 2023 , 89 : 102289 .

HAN R , LIU X , LI B , et al . The mechanism of the cavitation flow's spatiotemporal evolution and flow-induced instability [J ] . Flow Measurement and Instrumentation , 2025 , 104 : 102922 .

JIN Z , QIU C , JIANG C , et al . Effect of valve core shapes on cavitation flow through a sleeve regulating valve [J ] . Journal of Zhejiang University-SCIENCE A: Applied Physics & Engineering , 2020 , 21 ( 1 ): 1 - 14 .

干瑞彬 . 煤直接液化热高分液控阀失效分析 [J ] . 阀门 , 2020 , 4 : 17 - 22 .

GAN R . Failure analysis of the hydraulic control valve in coal direct liquefaction high temperature and pressure separator [J ] . Chinese Journal of Valve , 2020 , 4 : 17 - 22 .

刘启航 , 刘秀梅 , 李贝贝 , 等 . 调节阀内空化场压力累积强度特性分析 [J ] . 液压与气动 , 2023 , 47 ( 11 ): 113 - 119 .

LIU Q , LIU X , LI B , et al . Characteristic analysis of pressure cumulative strength of cavitation in control valve [J ] . Chinese Hydraulics & Pneumatics , 2023 , 47 ( 11 ): 113 - 119 .

刘忠 , 李显伟 , 邹淑云 , 等 . 基于混沌理论和CNN-OSVM的水轮机空化状态识别方法 [J ] . 动力工程学报 , 2023 , 43 ( 11 ): 1454 - 1460 .

LIU Z , LI X , ZOU S , et al . State recognition method of hydraulic turbine cavitation based on chaos theory and CNN-OSVM [J ] . Journal of Chinese Society of Power Engineering , 2023 , 43 ( 11 ): 1454 - 1460 .

刘忠 , 周泽华 , 邹淑云 , 等 . 基于迁移学习和CNN-LSTM的水轮机空化状态识别方法 [J ] . 动力工程学报 , 2024 , 44 ( 10 ): 1533 - 1540 .

LIU Z , ZHOU Z , ZOU S , et al . Cavitation state recognition method of hydraulic turbine based on transfer learning and CNN-LSTM [J ] . Journal of Chinese Society of Power Engineering , 2024 , 44 ( 10 ): 1533 - 1540 .

刘忠 , 郑佳稳 , 邹淑云 , 等 . 基于宽度卷积神经网络的水轮机空化状态识别方法 [J ] . 动力工程学报 , 2025 , 45 ( 7 ): 985 - 991 .

LIU Z , ZHENG J , ZOU S , et al . Hydraulic turbine cavitation state recognition method based on broad convolutional neural network [J ] . Journal of Chinese Society of Power Engineering , 2025 , 45 ( 7 ): 985 - 991 .

李贝贝 , 孙深圳 , 刘秀梅 , 等 . 基于改进SVM算法的调节阀空化状态识别 [J ] . 机械工程学报 , 2023 , 59 ( 24 ): 312 - 322 .

LI B , SUN S , LIU X , et al . Cavitation state recognition of regulating valve based on improved SVM algorithm [J ] . Journal of Mechanical Engineering , 2023 , 59 ( 24 ): 312 - 322 .

WEI J , LIU X , LI B , et al . Cavitation state recognition for control valve using AlexNet-type neural networks with three-channel images transformed by time series [J ] . Measurement Science and Technology , 2023 , 34 ( 5 ): 055301 .

邹淑云 , 刘忠 , 王文豪 , 等 . 基于分形维数和BiLSTM的离心泵空化状态识别方法 [J ] . 振动与冲击 , 2025 , 44 ( 04 ): 305 - 312 .

ZOU S , LIU Z , WANG W , et al . Cavitation recognition method of centrifugal pump based on fractal dimension and BiLSTM [J ] . Journal of Vibration and Shocks , 2025 , 44 ( 04 ): 305 - 312 .

LANG T , NI C , CHEN K , et al . Recognition of cavitation characteristics in non-clogging pumps based on the improved Lévy flight bat algorithm [J ] . Frontiers in Energy Research , 2023 , 11 : 1335227 .

OSTERMAN A , HOČEVAR M , ŠIROK B , et al . Characterization of incipient cavitation in axial valve by hydrophone and visualization [J ] . Experimental Thermal and Fluid Science , 2008 , 33 ( 4 ): 620 - 629 .

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 Y , MING T , HE G , et al . Cavitation state recognition of centrifugal pump based on deep learning [J ] . Journal of Xi’an Jiaotong University , 2017 , 51 ( 11 ): 165 - 172 .

YANG B , HWANG W , KO M , et al . Cavitation detection of butterfly valve using support vector machines [J ] . Journal of Sound and Vibration , 2004 , 287 ( 1 ): 25 - 43 .

PANDA K , RAPUR 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 .

石睿捷 , 宋戈 , 付豪 , 等 . 基于压力脉动分析的节流空化流场特性分析 [J ] . 航天制造技术 , 2023 , ( 4 ): 46 - 50 .

SHI R , SONG G , FU H , et al . Characteristic analysis of throttling cavitation flow field based on pressure fluctuation analysis [J ] . Aerospace Manufacturing Technology , 2023 , ( 4 ): 46 - 50 .

贺国 , 曹玉良 , 明廷锋 , 等 . 基于改进倍频带特征的离心泵空化状态识别 [J ] . 哈尔滨工程大学学报 , 2017 , 38 ( 8 ): 1263 - 1267+1302 .

HE G , CAO Y , MING T , et al . Cavitation state recognition of centrifugal pump based on features of modified octave bands [J ] . Journal of Harbin Engineering University , 2017 , 38 ( 8 ): 1263 - 1267+1302 .

SHA Y , FABER J , GOU S , et al . An acoustic signal cavitation detection framework based on XGBoost with adaptive selection feature engineering [J ] . Measurement , 2022 , 192 : 110897 .

ZHANG Y , LIU X , LI B , et al . Experiment investigation on cavitation bubble and mechanism of anti-cavitation performance in the bionic pressure relief valve [J ] . Proceedings of the Institution of Mechanical Engineers , Part E: Journal of Process Mechanical Engineering, 2023 : 09544089231214652 .

偶国富 , 饶杰 , 章利特 , 等 . 煤液化高压差调节阀空蚀/冲蚀磨损预测 [J ] . 摩擦学学报 , 2013 , 33 ( 2 ): 155 - 161 .

OU G , RAO J , ZHANG L , et al . Numerical investigation of cavitation erosion/solid particle erosion in high differential pressure control valves in coal liquefaction [J ] . Tribology , 2013 , 33 ( 2 ): 155 - 161 .

吴思宇 . 调节阀内空化流场动态演变特性研究及状态识别方法 [D ] . 徐州 : 中国矿业大学 , 2025 .

WU S . Study on dynamic evolution characteristics and state identification method of cavitation flow field in regulating valve [D ] . Xuzhou : China University of Mining and Technology , 2025 .

LI B , WU S , LIU X , et al . Experimental study on spatiotemporal distribution characteristics of different cavitation stages based on the flow deviation method [J ] . Proceedings of the Institution of Mechanical Engineers , Part E: Journal of Process Mechanical Engineering, 2024 : 09544089241303297 .

张金珂 , 张建刚 . 基于改进粒子群优化算法的信号检测及故障诊断 [J ] . 山东大学学报(理学版) , 2023 , 58 ( 05 ): 63 - 75+83 .

ZHANG J , ZHANG J . Signal detection and fault diagnosis based on improved particle swarm optimization algorithm [J ] . Journal of Shandong University (Natural Science) , 2023 , 58 ( 05 ): 63 - 75+83 .

TANGIRALA S . Evaluating the impact of GINI index and information gain on classification using decision tree classifier algorithm* [J ] . International Journal of Advanced Computer Science and Applications (IJACSA) , 2020 , 11 ( 2 ): 612 - 619 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用置信度分层协同决策的目标检测方法

机器学习在沸腾传热过程中的应用进展

跨声速压气机转子通用k-ω湍流模型的机器学习辅助优化

面向CO2捕集的吸附剂研究：从传统设计到机器学习

采用级联回归残差校正麻雀搜索算法优化长短期记忆网络的光伏发电功率预测方法