Two-phase flow pattern identification using continuous hidden Markov model

This paper presents a new method for identifying two-phase flow regimes from the instantaneous local fluid phase signals using continuous hidden Markov model (CHMM). CHMM is known to be a very strong pattern identifier. Air–water two-phase flows were realized in a transparent vertical tube. The tube length was 2 m, and its inner diameter was 19 mm. The instantaneous local fluid phase signals were collected using a single step index multimode optical fiber probe located at the center and mid-length of the tube. Signal features required in CHMM implementation were extracted using an innovative method. Various aspects of hidden Markov modeling and their effects on the results were studied. The flow pattern results are in very good agreement with photographs of the flow captured during the experiments. In sum, the results show that hidden Markov model has a good potential in identifying two-phase flow patterns.