Early Detection of Pump Failures
Innovate UK KTP Project (KTP ID #12028), ICT Escalator #567 and IMAGE #UB009.
2019-2023
Centrifugal pumps represent 70% of all kinds of pumps and are ubiquitous in the industrial world especially in heating, air conditioning and sewage applications. Although modern pumps can last for many years, their sudden failure can lead to undesirable or catastrophic disruptions.
The purpose of the project is to develop a low-cost IoT based predictive maintenance solution to continuously monitor the pump health using Motor current signature analysis (MCSA) and predict failures using a combination of signal processing and machine-learning algorithms. An MCSA monitoring system is deployed by attaching current clamps, used as transducers, to power supply wires without requiring direct physical access to the pump itself. The proposed system consists of custom hardware modules that stream the pump data to the cloud, and a backend for storage, visualisation and intelligent analysis. The project is a collaboration with
Team
|
听 |
Cem Ekin Sunal Marcus Bennett |
Barry Newton Jake Newton 听 听 |
听
Collaborators
| 听 |  |
Publications
- C. E. Sunal, V. Dyo and V. Velisavljevic, "Review of Machine Learning Based Fault Detection for Centrifugal Pump Induction Motors," in IEEE Access, vol. 10, pp. 71344-71355, 2022, doi: .
- C. E. Sunal, V. Velisavljevic, V. Dyo, B. Newton, J. Newton, 鈥淐entrifugal Pump Fault Detection with Convolutional Neural Network Transfer Learning鈥, in MDPI Sensors, vol. 24, no. 8, doi: 10.3390/s24082442
Patent
- C. E. Sunal, V. Velisavljevic, V. Dyo, B. Newton, J. Newton, 鈥淔ault detection and monitoring for electric pump motors鈥, Patent file GB2402869.8, pending.
听