Using Digital Models for Condition Based Maintenance of High Pressure Pumps in SWRO Desalination Plants

Abstract

Condition based maintenance (CBM) is a very useful technique built around the concept of monitoring the operation of assets. This allows performing diagnostics, troubleshooting and determining the proper maintenance action. Conventional CBM programs are solely built around analyzing sensors’ data collected from physical systems. Incorporating simulation data collected from digital replica of the system with sensors’ data can lead to more optimization for operation and maintenance. This paper focuses on demonstrating the role of digital models in implementing effective condition based maintenance on high pressure pumps in SWRO desalination plants. Two digital models have been developed for the high pressure pump experiencing cavitation condition. The first digital model is a 3-D fully turbulent Computational Fluid Dynamic (CFD) model. Based on the pressure distribution obtained from the CFD model, pressure pulsation due to impeller-diffuser interaction is studied and used to analytically model the exciting forces affecting the rotor. The second digital model is a pump rotordynamic model which is used to predict the rotor vibration response to exciting forces at both normal and faulty operating conditions. Results obtained from the digital models are validated using an experimental test rig of a small centrifugal pump. Using this concept, a pump faulty operation can be simulated to provide complete understanding of the root cause of the fault. Additionally, digital models can be used to simulate different corrective actions that would restore the normal operation of the pump.