Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Condition monitoring of a centrifugal pump using Bayesian networks

Kapuria, Abhimanyu (2023) Condition monitoring of a centrifugal pump using Bayesian networks. Master's Thesis, University of Pittsburgh. (Unpublished)

Download (2MB) | Preview


To improve the viability of nuclear power plants, there is a need to reduce their operational costs. Operational costs account for a significant portion of a nuclear plant's yearly budget. This is due to their reactive and uninformed maintenance approach. In order to reduce these costs, proactive condition monitoring methods are required that can estimate the state of a machine in real-time and aid operators in optimizing maintenance schedules. In this research, we use Bayesian networks to develop a condition monitoring platform that can diagnose pump faults, infer their root cause, and estimate it's remaining-useful-life. This solution provides an informed probabilistic viewpoint of the pumping system for the purpose of preventative maintenance.

In this thesis, we analyze a centrifugal pump to estimate its current state. We combine domain expertise with physical laws to create a cause-and-effect relationship between the pump components to estimate the root cause of two major pump faults. We incorporate survival analysis with Bayesian statistics to forecast the health of the pump conditional to its current state. We complete our condition monitoring analysis by successfully implementing the final Bayesian network in case studies for three modes of pump operation.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Kapuria, Abhimanyuabk80@pitt.eduabk800000000220962771
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Thesis AdvisorCole, Daniel
Committee MemberClark,
Committee MemberBajaj,
Date: 13 June 2023
Date Type: Publication
Defense Date: 27 July 2022
Approval Date: 13 June 2023
Submission Date: 24 January 2023
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 88
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Mechanical Engineering
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: condition monitoring, machine learning, fault detection, remaining useful life, survival analysis, Bayesian networks, root cause, probabilistic estimation, vibration analysis, machine health
Date Deposited: 13 Jun 2023 14:03
Last Modified: 13 Jun 2023 14:03


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item