Pattern recognition in flow visualization around a paddle impeller

Sugeng Winardi, Shuya Nakao, Yoichi Nagase

Research output: Contribution to journalArticlepeer-review

27 Citations (Scopus)


Very low frequency fluctuations in the velocity signal from several seconds to several minutes in scale are usually observed in agitated vessels. It is shown in this work that these fluctuations result from the abrupt replacement of the flow pattern in the vessel, even in part. Pattern recognition studies of the flow in the impeller region of a paddle were carried out, using high-speed VTR. The results showed that the flow between blades is characterized into five patterns which are replaced one after another. They are classified into the discharge and the cross-pass types of flow. The latter means that flow passes between impeller blades, upwardly or downwardly. The lifetime of the discharge flow type was three times that of the cross-pass flow type. The velocity distributions of the discharge and cross-pass flow types estimated from visualized pictures were quite different from each other and were different from the average distributions obtained by LDV. When the cross-pass pattern had developed over the whole region between blades, a horizontal figure-eight pattern of flow developed in the entire vessel with abrupt reduction of the torque.

Original languageEnglish
Pages (from-to)503-508
Number of pages6
JournalJournal of Chemical Engineering of Japan
Issue number5
Publication statusPublished - 1988
Externally publishedYes


  • Conditional Sampling
  • Flow Recognition Impeller Stream
  • Mixing
  • Paddle Impeller
  • Torque Fluctuations
  • Velocity Distributions
  • Visualization


Dive into the research topics of 'Pattern recognition in flow visualization around a paddle impeller'. Together they form a unique fingerprint.

Cite this