Improved CA-ResNet Network for Residual Life Prediction of 17-4PH Material Damaged by Cavitation Erosion

DI Juan; WANG Chengbo; HE Lei; NIU Yalong; PENG Chaoyi; WANG Jianfeng

doi:10.7652/xjtuxb202501019

您当前的位置：

首页 >

文章列表页 >

Improved CA-ResNet Network for Residual Life Prediction of 17-4PH Material Damaged by Cavitation Erosion

更新时间：2026-01-06

- Improved CA-ResNet Network for Residual Life Prediction of 17-4PH Material Damaged by Cavitation Erosion
- Journal of Xi’an Jiaotong University Vol. 59, Issue 1, Pages: 206-214(2025)
- 作者机构：
  
  1.太原科技大学车辆与交通工程学院， 030024，太原
  2.株洲时代新材料科技股份有限公司， 412007，湖南株洲
  3.湖南大学材料科学与工程学院， 410082，长沙
  4.同济大学工程结构性能演化与控制教育部重点实验室， 200092，上海
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202501019
  CLC： TK405
- Received：29 May 2024，
  
  Online First：19 July 2024，
  
  Published：10 January 2025
- 稿件说明：
移动端阅览
DI Juan, WANG Chengbo, HE Lei, et al. Improved CA-ResNet Network for Residual Life Prediction of 17-4PH Material Damaged by Cavitation Erosion[J]. Journal of Xi’an Jiaotong University, 2025, 59(1): 206-214.
DOI：

DI Juan, WANG Chengbo, HE Lei, et al. Improved CA-ResNet Network for Residual Life Prediction of 17-4PH Material Damaged by Cavitation Erosion[J]. Journal of Xi’an Jiaotong University, 2025, 59(1): 206-214. DOI： 10.7652/xjtuxb202501019.

摘要

为解决材料汽蚀损伤后剩余使用寿命预测困难的科学问题，提出了一种融合卷积神经网络技术的寿命预测方法，具体采用残差网络（ResNet）模型，并嵌入坐标注意力机制（CA），通过对模型进行卷积、通道数及下采样方式的优化，构建改进的坐标残差网络（CA-ResNet）模型，以实现对17-4PH材料汽蚀损伤后剩余使用寿命的精确预测。基于超声波汽蚀试验得到的汽蚀特性曲线，通过逻辑回归（Logistic）方程对汽蚀阶段进行定量划分，并定义寿命系数

，同时借助超景深显微镜获取材料损伤后不同时刻的显微图像数据库，并与寿命系数

相对应。研究结果表明，改进的CA-ResNet网络模型在CIFAR10公开数据集上验证准确率可达92.2%，在收集的17-4PH材料的汽蚀损伤数据集上的验证准确率可达93.2%，相比ResNet18网络模型，准确率分别提高了1.5%和3.5%。通过学习率、批处理量等超参数优化后，该模型在汽蚀损伤数据集上准确率可达95.0%。采用端到端的数据驱动思想，可实现从汽蚀损伤形貌到汽蚀寿命的精确预测。

Abstract

In order to address the scientific challenge of predicting the remaining useful life of materials damaged by cavitation erosion

a life prediction method that integrates convolutional neural network technology is introduced. Specifically

the residual network (ResNet) model is utilized and enhanced with a coordinate attention mechanism (CA). Through optimization of the convolution

channel number and down-sampling method in the model

an improved coordinate residual network (CA-ResNet) model is developed to accurately predict the remaining useful life of 17-4PH material damaged by cavitation erosion. Initially

the cavitation characteristic curve is obtained from ultrasonic cavitation tests. Then

the cavitation stages are quantitatively segmented using a Logistic equation

defining the life coefficient

. Meanwhile

with the assistance of a super-depth-of-field microscope

microscopic images of the material at various post-damage time points are captured to establish a microscopic image database correlated with the life coefficient

. The results demonstrate that the improved CA-ResNet network model achieves a verification accuracy of 92.2% on the CIFAR10 public dataset and 93.2% on the collected cavitation damage dataset of 17-4PH material. This represents a 1.5% and 3.5% accuracy improvement over the ResNet18 network model

respectively. By fine-tuning hyperparameters like learn

ing rate and batch size

the accuracy on the cavitation damage dataset is elevated to 95.0%. In this paper

an end-to-end data-driven approach is adopted to achieve accurate prediction from cavitation damage morphology to post-cavitation life.

关键词

Keywords

references

李大勇 , 郭维民 , 唐洪岩 . 离心泵的汽蚀问题与防止措施 [J ] . 南方农机 , 2020 , 51 ( 14 ): 43 - 44 .

LI Dayong , GUO Weimin , TANG Hongyan . Cavitation problem and prevention measures of centrifugal pumps [J ] . China Southern Agricultural Machinery , 2020 , 51 ( 14 ): 43 - 44 .

王勇 , 刘厚林 , 谈明高 . 泵汽蚀研究现状及展望 [J ] . 水泵技术 , 2008 ( 1 ): 1 - 4 .

WANG Yong , LIU Houlin , TAN Minggao . Current status and prospects of pump cavitation research [J ] . Pump Technology , 2008 ( 1 ): 1 - 4 .

DULAR M , STOFFEL B , ŠIROK B . Development of a cavitation erosion model [J ] . Wear , 2006 , 261 ( 5/6 ): 642 - 655 .

LAGUNA-CAMACHO J R , LEWIS R , VITE-TORRES M , et al . A study of cavitation erosion on engineering materials [J ] . Wear , 2013 , 301 ( 1/2 ): 467 - 476 .

FERRARI A . Fluid dynamics of acoustic and hydrodynamic cavitation in hydraulic power systems [J ] . Proceedings of the Royal Society: A Mathematical, Physical and Engineering Sciences , 2017 , 473 ( 2199 ): 20160345 .

ADAMKOWSKI A , HENKE A , LEWANDOWSKI M . Resonance of torsional vibrations of centrifugal pump shafts due to cavitation erosion of pump impellers [J ] . Engineering Failure Analysis , 2016 , 70 : 56 - 72 .

FONTANESI S , GIACOPINI M , CICALESE G , et al . Numerical investigation of the cavitation damage in the wet cylinder liner of a high performance motorbike engine [J ] . Engineering Failure Analysis , 2014 , 44 : 408 - 423 .

廖乃坚 , 谢代标 . 关于LJ465Q系列发动机水泵气蚀的探讨 [J ] . 装备制造技术 , 2010 ( 2 ): 177 - 179 .

LIAO Naijian , XIE Daibiao . Discussion on the pump cavitation in LJ465Q Series Engine [J ] . Equipment Manufacturing Technology , 2010 ( 2 ): 177 - 179 .

王瑞杰 . 内燃机的冷却水泵的汽蚀工况分析和提高汽蚀性能设计方法的探讨 [J ] . 淮海工学院学报(自然科学版) , 1996 , 5 ( 1 ): 19 - 25 .

WANG Ruijie . Discussion on the working condition of cavitation and improvement of design of the cooling water pump in internal combustion engine [J ] . Journal of Huaihai Institute of Technology (Natural Sciences Edition) , 1996 , 5 ( 1 ): 19 - 25 .

彭健 , 何世权 . 高压调节阀结构改进与汽蚀仿真 [J ] . 液压与气动 , 2019 ( 3 ): 120 - 125 .

PENG Jian , HE Shiquan . Structural improvement and cavitation simulation of high pressure regulating valve [J ] . Chinese Hydraulics & Pneumatics , 2019 ( 3 ): 120 - 125 .

邸娟 , 王顺森 , 蒋希航 , 等 . 透平末级叶片材料抗水蚀特性的数值研究 [J ] . 西安交通大学学报 , 2021 , 55 ( 2 ): 38 - 46 .

DI Juan , WANG Shunsen , JIANG Xihang , et al . Numerical research on water erosion resistance characteristics of the substrate material of last stage blades in steam turbine [J ] . Journal of Xi'an Jiaotong University , 2021 , 55 ( 2 ): 38 - 46 .

刘春燕 , 李敬民 . 17-4PH沉淀硬化不锈钢力学性能稳定性研究 [J ] . 热加工工艺 , 2018 , 47 ( 20 ): 178 - 181 .

LIU Chunyan , LI Jingmin . Research on mechanical properties stability of 17-4PH precipitated hardened stainless steel [J ] . Hot Working Technology , 2018 , 47 ( 20 ): 178 - 181 .

薛承感 , 潘慧斌 , 丁银萍 , 等 . 17-4PH汽轮机叶片激光固溶强化抗汽蚀性能研究 [J ] . 动力工程学报 , 2020 , 40 ( 1 ): 82 - 88 .

XUE Chenggan , PAN Huibin , DING Yinping , et al . Cavitation erosion behavior of 17-4PH steam turbine blades treated by laser solid solution [J ] . Journal of Chinese Society of Power Engineering , 2020 , 40 ( 1 ): 82 - 88 .

DULAR M , COUTIER-DELGOSHA O . Numerical modelling of cavitation erosion [J ] . International Journal for Numerical Methods in Fluids , 2009 , 61 ( 12 ): 1388 - 1410 .

OSTERMAN A , BACHERT B , SIROK B , et al . Time dependent measurements of cavitation damage [J ] . Wear , 2009 , 266 ( 9/10 ): 945 - 951 .

SOYAMA H , FUTAKAWA M . Estimation of incubation time of cavitation erosion for various cavitating conditions [J ] . Tribology Letters , 2004 , 17 ( 1 ): 27 - 30 .

HATTORI S , HIROSE T , SUGIYAMA K . Prediction method for cavitation erosion based on measurement of bubble collapse impact loads [J ] . Wear , 2010 , 269 ( 7/8 ): 507 - 514 .

AL-OBAIDI A R . Detection of cavitation phenomenon within a centrifugal pump based on vibration analysis technique in both time and frequency domains [J ] . Experimental Techniques , 2020 , 44 ( 3 ): 329 - 347 .

CAO Ruijia , YUAN Jianping , DENG Fanjie , et al . Numerical method to predict vibration characteristics induced by cavitation in centrifugal pumps [J ] . Measurement Science and Technology , 2021 , 32 ( 11 ): 115109 .

CDINA M . Detection of cavitation phenomenon in a centrifugal pump using audible sound [J ] . Mechanical Systems and Signal Processing , 2003 , 17 ( 6 ): 1335 - 1347 .

HE Kaiming , ZHANG Xiangyu , REN Shaoqing , et al . Deep residual learning for image recognition [C ] // 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway, NJ, USA : IEEE , 2016 : 770 - 778 .

HE Kaiming , ZHANG Xiangyu , REN Shaoqing , et al . Identity mappings in deep residual networks [C ] // Computer Vision-ECCV 2016 . Cham, Switzerland : Springer International Publishing , 2016 : 630 - 645 .

ZHANG Zheyuan , LIU Tianyuan , ZHANG Di , et al . Water droplet erosion life prediction method for steam turbine blade materials based on image recognition and machine learning [J ] . Journal of Engineering for Gas Turbines and Power , 2021 , 143 ( 3 ): 031009 .

HU Jie , SHEN Li , SUN Gang . Squeeze-and-excitation networks [C ] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway, NJ, USA : IEEE , 2018 : 7132 - 7141 .

HOU Qibin , ZHOU Daquan , FENG Jiashi . Coordinate attention for efficient mobile network design [C ] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway, NJ, USA : IEEE , 2021 : 13708 - 13717 .

HOWARD A G , ZHU Menglong , CHEN Bo , et al . MobileNets: efficient convolutional neural networks for mobile vision applications [EB/OL ] . ( 2017-04-17 ) [ 2024-04-19 ] . https://arxiv.org/abs/1704.04861 https://arxiv.org/abs/1704.04861 .

KRIZHEVSKY A . Learning multiple layers of features from tiny images [EB/OL ] . ( 2009-04-08 ) [ 2024-04-13 ] . http://www.cs.toronto.edu/~kriz/learning-features-2009-TR.pdf http://www.cs.toronto.edu/~kriz/learning-features-2009-TR.pdf .

DI Juan , WANG Shunsen , YAN Xiaojiang , et al . Experimental investigation on effect of surface strengthening process and roughness on water droplet erosion behavior in turbomachinery [J ] . Tribology International , 2021 , 153 : 106647 .

YANG Li , SHAMI A . On hyperparameter optimization of machine learning algorithms: theory and practice [J ] . Neurocomputing , 2020 , 415 : 295 - 316 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Influence of Structural Parameters on the Summer Operational Characteristics of a Three-Tower-in-One Indirect Air-Cooling System

Bi-Level Multi-Obj ective Optimization of a Wind-Solar Coupled Hydrogen Production System Based on the Synergy of Hybrid Electrolyzer Arrays

Mechanism and Optimal Control Strategy for Hysteresis in Rapid Dehumidification Response of Indirect Thermal Systems in Electric Vehicles

Numerical Simulation of the Energy Conversion Process in Thermoacoustic Engines with Different Stack Materials

Windage Loss and Flow Characteristics in Impeller Back Gaps of sCO2 Radial Inflow Turbines

Related Author

No data

Related Institution

No data

⁰