Impact of gas-liquid two-phase flow patterns in the suction chamber of a centrifugal pump on performance efficiency

• The statistical characteristics were examined based on the air-liquid flow parameters. • The characteristic statistical methodology and wavelet analysis can distinguish flow patterns. • Identified flow patterns correlate with pump performance characteristics. • Agglomerated flow leads to pump surging, risking damage and performance loss. As a result of the degradation of pump performance, the impact of various two-phase flow patterns in the suction line of a centrifugal pump owing to the varying air and water flow rate is crucial to understand. The purpose of this study is to conduct a statistical analysis of the time variation of the pressure gradient data to distinguish various flow patterns in the suction side of the pump on its performance. Furthermore, the study identifies the specific characteristics of the reduction in pump performance reduction associated with each flow pattern. For time-domain, frequency-domain, and time–frequency analyses, statistical diagnostic tools are used. Those include the probability density function (PDF), the power spectral density (PSD), continuous wavelet transforms, and Kolmogorov entropy. The results show that the combinations of statistical signal characterization, wavelet analysis, and Kolmogorov entropy analysis can distinguish flow patterns and transitions generated in the suction line of a centrifugal pump. This research identifies three distinct flow patterns and discovers a correlation between them and dimensionless pump characteristics. Additionally, an indicator of considerable deterioration in the pump head is indicated by the occurrence of surging due to the entry of agglomerated flow.