A new methodology was developed for flow regime identification in pipes.The method utilizes the pattern recognition abilities of Artificial Neural Networksand the unprocessed time series of a system-monitoring-signal.The methodology was tested with synthetic data from a conceptual system,liquid level indicating Capacitance signals from a Horizontal flow systemand with a pressure difference signal from a S-shape riser.The results showed that the signals that were generated for the conceptualsystem had all their patterns identified correctly with no errors what so ever.The patterns for the Horizontal flow system were also classified very wellwith a few errors recorded due to original misclassifications of the data. Themisclassifications were mainly due to subjectivity and due to signals thatbelonged to transition regions, hence a single label for them was not adequate.Finally the results for the S-shape riser showed also good agreement with thevisual observations and the few errors that were identified were again due tooriginal misclassifications but also to the lack of long enough time series forsome flow cases and the availability of less flow cases for some flow regimesthan others.In general the methodology proved to be successful and there were anumber of advantages identified for this neural network methodology in comparisonto other ones and especially the feature extraction methods. Theseadvantages were: Faster identification of changes to the condition of thesystem, inexpensive suitable for a variety of pipeline geometries and morepowerful on the flow regime identification, even for transitional cases.
展开▼
