Wp. Liu et Cs. James, Estimation of discharge capacity in meandering compound channels using artificial neural networks, CAN J CIV E, 27(2), 2000, pp. 297-308
Flow in compound (or two-stage) channels is very complex and different ener
gy loss mechanisms operate under different geometric and flow conditions. N
either theoretical analyses nor current empirical approaches are sufficient
ly developed for practical calculation of conveyance for all conditions exp
erienced in practice. An alternative approach, using artificial neural netw
ork modelling, has been successfully applied to predict conveyance under a
wide range of conditions. The model proposed uses a feed-forward system wit
h one hidden layer and an error back-propagation learning procedure. It pre
dicts a dimensionless discharge using input describing the main channel and
floodplain flow depths, vegetation density over the cross section, channel
sinuosity, transverse floodplain slope, and floodplain bend tightness. The
discharge is dimensionalized by multiplication with the composite discharg
e calculated assuming frictional resistance only. The model was trained usi
ng 45 data sets representing a range of main channel and floodplain charact
eristics and tested using 15 additional data sets. The discharge prediction
error for all the data used in development and testing the model was -0.19
% on average and exceeded 15% for one condition only.