By: G. S. Dave, A. P. Pandhare, Abhishek Kamane, Ajinkya Konkar Konkar, Atharva Konde, and Sahil Mane
1. PHD Scholar, Sinhgad College of Engineering, Department of Mechanical Engineering, Assistant Professor, Smt. Kashibai Navale College Engineering (SKNCOE), Vadgaon Budruk,
Pune, Maharashtra, India
2. Professor, Department of Mechanical Engineering, Professor, Sinhgad College of Engineering (SCOE), Vadgaon Budruk, Pune, Maharashtra, India
3-6 Student, Department of Mechanical Engineering, Bachelors in Engineering, Smt Kashibai Navale College of Engineering (SKNCOE), Vadgaon Budruk, Pune, Maharashtra, India
Centrifugal pumps are crucial equipment widely used in various industries, and their reliable operation is essential for maintaining production and preventing costly downtimes. This paper proposes a comprehensive health monitoring system for centrifugal pumps that combines model-driven and data- driven approaches, leveraging artificial intelligence (AI) and machine learning (ML) techniques. The system incorporates demand prediction using artificial neural networks and fault diagnosis employing different ML classifiers, such as decision trees and logistic regression. The model-driven approach involves reconstructing the pump dynamics from vibration signals and analyzing the phase space for fault detection. The data-driven methods utilize support vector machines, k-nearest neighbours, and other algorithms for fault diagnosis based on acquired data. The proposed system aims to provide real-
time, comprehensive information to management, enabling predictive maintenance and minimizing the risk of equipment failure. The paper presents a comparison of different classifiers and health monitoring systems, highlighting the advantages of the proposed comprehensive model. The results demonstrate the effectiveness of the combined decision tree and logistic regression algorithms as a robust foundation for centrifugal pump health monitoring. Additionally, the paper explores the integration of Internet of Things (IoT) and edge computing for real-time monitoring and analysis, addressing challenges such as data security and privacy.
Keywords: Centrifugal Pump, Artificial Intelligence, Machine Learning, Data Acquisition, Internet of
Things
Citation:
Refrences:
- L. Costa Brito, G. Antonio Susto, J. Nei Brito, and M. Antonio Viana Duarte, “AnExplainable
Artificial Intelligence Approach for Unsupervised Fault - H. M. Chen, J. H. Zhang, Y. C. Wang, H. C. Chang et al., “Hot-Pressing Furnace Current
Monitoring and Predictive Maintenance System in Aerospace Applications,” 2023. - S. F. Chevtchenko, M. C. M. dos Santos, D. M. Vieira, R. L. Mota et al., “PredictiveMaintenance
Model Based on Anomaly Detection in Induction Health Monitoring System for Fault Identification Dave Et All - Motors: A Machine Learning Approach Using Real-Time IoT Data,” 2023.
- Y. Yang, L. Ding, J. Xiao, G. Fang et al., “Current Status and Applications for HydraulicPump
Fault Diagnosis: A Review,” 2022. - Y. Luo, Y. Han, S. Yuan, and J. Yuan, “Research on the Single-Value Indicators forCentrifugal
Pump Based on Vibration Signals,” 2020. - M. Irfan et al., “A Novel Feature Extraction and Fault Detection Technique for the Intelligent
Fault Identification of Water Pump Bearings,” Sensors, vol. 21, no. 12, p. 4225,Jun. 2021, doi:
10.3390/s21124225. - Y. Xiao, Y. Li, and C. Chu, “Performance Analysis of Vibration Sensors for Closed-Loop
Feedback Health Monitoring of Mechanical Equipment,” J. Sens., vol. 2021, pp. 1– 12, Oct.
2021, doi: 10.1155/2021/6348347. - G. Mousmoulis, N. Karlsen-Davies, G. Aggidis, I. Anagnostopoulos, and D. Papantonis,
“Experimental analysis of cavitation in a centrifugal pump using acoustic emission, vibration
measurements and flow visualization,” Eur. J. Mech. – BFluids, vol. 75, pp. 300–311, May
2019, doi: 10.1016/j.euromechflu.2018.10.015. - M. Karami and R. Madlener, “Smart Predictive Maintenance Strategy Based on Cyber-Physical
Systems for Centrifugal Pumps: A Bearing Vibration Analysis,” SSRN Electron. J., 2019, doi:
10.2139/ssrn.3544199. - M. Abd-Elaal, M. Lotfy, T. Nabil, and A. Eldomiaty, “Condition-based monitoring of a small
centrifugal pump by vibration analysis,” IOP Conf. Ser. Mater. Sci. Eng., vol. 610, no. 1, p.
012049, Sep. 2019, doi: 10.1088/1757-899X/610/1/012049. - J. S. Rapur and R. Tiwari, “Experimental fault diagnosis for known and unseen operating
conditions of centrifugal pumps using MSVM and WPT based analyses,”Measurement, vol.
147, p. 106809, Dec. 2019, doi: 10.1016/j.measurement.2019.07.037. - Z. Ahmad, T.-K. Nguyen, S. Ahmad, C. D. Nguyen, and J.-M. Kim, “Multistage Centrifugal
Pump Fault Diagnosis Using Informative Ratio Principal Component Analysis,” Sensors, vol.
22, no. 1, p. 179, Dec. 2021, doi: 10.3390/s22010179.
