| Abstrak/Abstract |
In the oil and gas industry, centrifugal pump operation with gas-liquid mixtures
is a real concern. The main problem is the ingress of undesired air into the centrifugal pump
which can decrease pump performance. In addition, the pressure transmitters installed at the
pump inlet and outlet cannot identify the two-phase flow pattern formed at the inlet and
outlet of the pump. This study aims to identify the correlation pressure signal characteristic
between the inlet and outlet pump and the effect of flow pattern at the suction pump on
suction and discharge pressure. Superficial air velocity is varied JG, in= 0.02, 0.03, 0.05, 0.07,
0.08, 0.1 and 0.12 m/s and injected to the inlet pump. Pressure transmitters are mounted at
suction and discharge of centrifugal pump to record pressure fluctuations and a 1-inch clear
transparent pipe is installed at the upstream pump to ease flow pattern visualization. The
fluctuation pressure signal data is analyzed using time series, fast fourier transform (FFT),
probability density function (PDF) and power spectral density (PSD). In addition, the
recorded pressure fluctuations are used to observe their effect on the flow pattern formed at
the pump inlet. The results show that he visualization analysis revealed four distinct flow
patterns in the suction chamber: bubble flow, agglomerated bubble flow, wavy flow, and
stratified flow which the flow pattern change affects the signal time series, FFT, PDF, and
PSD. The flow at the suction side of the pump will be pulsating when air is injected from
0.02 m/s to 0.08m/s with several flow patterns visible at each condition. Frequency dominant
at FFT diagram and PSD diagram shows a consistent number between inlet and outlet pump.
However, superficial air velocity of 0.1 m/s and 0.12 m/s generates stratified wavy and
stratified flows, making the pump performance drop significantly |
