Browse By Repository:

 
 
 
   

Pump monitoring using multi-sensor and analysis by statistical technique

Awang Khalid, Dayang Nurhafiza (2024) Pump monitoring using multi-sensor and analysis by statistical technique. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

[img] Text (Full Text)
Pump monitoring using multi-sensor and analysis by statistical technique.pdf - Submitted Version

Download (2MB)

Abstract

Centrifugal pumps are widely used in both industrial and municipal water systems due to their efficiency and versatility. These pumps work by converting mechanical energy into kinetic energy to move the fluids, typically water, from one place to another. However, during their operation, centrifugal pumps can experience several failures or issues that can affect their performance and efficiency. In recent years, the development of online pump monitoring strategies has emerged as a valuable approach to detect faults and anomalies in pump systems. This project aims to investigate online pump monitoring utilizing vibration signal analysis and a machine learning method with a 2-type different sensor of accelerometer and piezoelectric film sensor. By analyzing the vibration signal generated during pump operation, this technique enables real-time assessment of pump performance and health. A piezoelectric film sensor is affixed to the pump to monitor and record the vibration signal produced by its components. The gathered information from the vibration signal is then transmitted for advanced signal processing. This involves analyzing the frequency, amplitude, and other characteristics of the vibration signals to extract crucial features that indicate the actual condition of the pump. Machine learning algorithms are employed to analyze the extracted features and provide historical data and patterns that can be utilized to predict the health status of the pump. The research results demonstrate that different pump speeds, valve opening, and flow rates generate distinct vibration signal patterns. This knowledge enhances our understanding of pump behavior, enabling better defect detection and ultimately leading to improvements in pump performance.

Item Type: Final Year Project (Project Report)
Uncontrolled Keywords: Centrifugal pumps, Vibration signal analysis, Machine learning, Online pump monitoring, Fault detection
Subjects: T Technology > T Technology (General)
T Technology > TA Engineering (General). Civil engineering (General)
Divisions: Library > Final Year Project > FTKM
Depositing User: Norfaradilla Idayu Ab. Ghafar
Date Deposited: 09 Dec 2024 09:15
Last Modified: 09 Dec 2024 09:15
URI: http://digitalcollection.utem.edu.my/id/eprint/32751

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year