Deep Learning for Enhanced Fault Diagnosis of Monoblock Centrifugal Pumps: Spectrogram-Based Analysis

被引:16
|
作者
Chennai Viswanathan, Prasshanth [1 ]
Venkatesh, Sridharan Naveen [1 ]
Dhanasekaran, Seshathiri [2 ]
Mahanta, Tapan Kumar [1 ]
Sugumaran, Vaithiyanathan [1 ]
Lakshmaiya, Natrayan [3 ]
Paramasivam, Prabhu [4 ]
Nanjagoundenpalayam Ramasamy, Sakthivel [5 ]
机构
[1] Vellore Inst Technol, Sch Mech Engn, Vandalur Kelambakkam Rd, Chennai 600127, India
[2] UiT Arctic Univ Norway, Dept Comp Sci, N-9037 Tromso, Norway
[3] Saveetha Sch Engn, SIMATS, Dept Mech Engn, Chennai 602105, India
[4] Mattu Univ, Coll Engn & Technol, Dept Mech Engn, Mettu 318, Ethiopia
[5] Amrita Vishwa Vidyapeetham, Amrita Sch Engn, Dept Mech Engn, Coimbatore 641112, India
关键词
monoblock centrifugal pumps; spectrogram; MEMS accelerometer; pre-trained networks; deep learning;
D O I
10.3390/machines11090874
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The reliable operation of monoblock centrifugal pumps (MCP) is crucial in various industrial applications. Achieving optimal performance and minimizing costly downtime requires effectively detecting and diagnosing faults in critical pump components. This study proposes an innovative approach that leverages deep transfer learning techniques. An accelerometer was adopted to capture vibration signals emitted by the pump. These signals are then converted into spectrogram images which serve as the input for a sophisticated classification system based on deep learning. This enables the accurate identification and diagnosis of pump faults. To evaluate the effectiveness of the proposed methodology, 15 pre-trained networks including ResNet-50, InceptionV3, GoogLeNet, DenseNet-201, ShuffleNet, VGG-19, MobileNet-v2, InceptionResNetV2, VGG-16, NasNetmobile, EfficientNetb0, AlexNet, ResNet-18, Xception, ResNet101 and ResNet-18 were employed. The experimental results demonstrate the efficacy of the proposed approach with AlexNet exhibiting the highest level of accuracy among the pre-trained networks. Additionally, a meticulous evaluation of the execution time of the classification process was performed. AlexNet achieved 100.00% accuracy with an impressive execution (training) time of 17 s. This research provides invaluable insights into applying deep transfer learning for fault detection and diagnosis in MCP. Using pre-trained networks offers an efficient and precise solution for this task. The findings of this study have the potential to significantly enhance the reliability and maintenance practices of MCP in various industrial settings.
引用
收藏
页数:17
相关论文
共 50 条
  • [1] Deep learning for fault diagnosis of monoblock centrifugal pumps: a Hilbert-Huang transform approach
    Prasshanth, C. V.
    Venkatesh, S. Naveen
    Mahanta, Tapan K.
    Sakthivel, N. R.
    Sugumaran, V.
    INTERNATIONAL JOURNAL OF SYSTEM ASSURANCE ENGINEERING AND MANAGEMENT, 2024,
  • [2] Fault diagnosis of monoblock centrifugal pumps using pre-trained deep learning models and scalogram images
    Prasshanth, Chennai Viswanathan
    Venkatesh, Sridharan Naveen
    Mahanta, Tapan Kumar
    Sakthivel, Nanjagoundenpalayam Ramasamy
    Sugumaran, Vaithiyanathan
    ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 136
  • [3] Emotional sounds of crowds: spectrogram-based analysis using deep learning
    Franzoni, Valentina
    Biondi, Giulio
    Milani, Alfredo
    MULTIMEDIA TOOLS AND APPLICATIONS, 2020, 79 (47-48) : 36063 - 36075
  • [4] Emotional sounds of crowds: spectrogram-based analysis using deep learning
    Valentina Franzoni
    Giulio Biondi
    Alfredo Milani
    Multimedia Tools and Applications, 2020, 79 : 36063 - 36075
  • [5] A Fault Diagnosis Framework for Centrifugal Pumps by Scalogram-Based Imaging and Deep Learning
    Hasan, Md Junayed
    Rai, Akhand
    Ahmad, Zahoor
    Kim, Jong-Myon
    IEEE ACCESS, 2021, 9 : 58052 - 58066
  • [6] Deep learning-based multilabel compound-fault diagnosis in centrifugal pumps
    Jiang, Lizhe
    Du, Hongze
    Bu, Yufeng
    Zhao, Chunyu
    Lu, Hailong
    Yan, Jun
    OCEAN ENGINEERING, 2024, 314
  • [7] Mel Spectrogram-based advanced deep temporal clustering model with unsupervised data for fault diagnosis
    Hong, Geonkyo
    Suh, Dongjun
    EXPERT SYSTEMS WITH APPLICATIONS, 2023, 217
  • [8] Fault Diagnosis for Centrifugal Pumps Using Deep Learning and Softmax Regression
    Zhao, Wanlin
    Wang, Zili
    Lu, Chen
    Ma, Jian
    Li, Lianfeng
    PROCEEDINGS OF THE 2016 12TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION (WCICA), 2016, : 165 - 169
  • [9] Audio Signal Mapping into Spectrogram-Based Images for Deep Learning Applications
    Ciric, Dejan
    Peric, Zoran
    Nikolic, Jelena
    Vucic, Nikola
    2021 20TH INTERNATIONAL SYMPOSIUM INFOTEH-JAHORINA (INFOTEH), 2020,
  • [10] Multi-kernel Learning based Autonomous Fault Diagnosis for Centrifugal Pumps
    Feng, Haoran
    Wang, Yongcai
    Zhu, Jiajun
    Li, Deying
    2018 INTERNATIONAL CONFERENCE ON CONTROL, AUTOMATION AND INFORMATION SCIENCES (ICCAIS), 2018, : 548 - 553