Velocity measurements in a concrete experimental channel representing a flood plain
Mesures de vitesse dans un canal expérimental représentant une plaine d'inondation
Résumé
Flows in compound channels are characterized by huge discharge and momentum transfers between the flood plain and the main channel. They occur with a strong conveyance modification (obstacle) or when the discharge ratio between the two beds is different from a uniform flow solution. An experimental channel was built in the Laboratory of Hydraulics of the Compagnie Nationale du Rhône. The 3 m by 13 meters channel was divided in two parallel beds, a 0.80 m wide main channel of trapezoidal shape and a 2.1m wide flood plain. The curvature radius was about 25 m at approximately half length of the channel. This location was chosen to introduce an obstacle in the flood plain which reduces the width by one third to two thirds. Determination of the Manning coefficients of both main channel and flood plain were performed by isolating one after the other channel. A different corrugation rose the value in the flood plain compared to the main channel. Water depth were acquired thanks to point gauges. The velocities at various depths (2 to 5) were measured by a small propeller which provides the value in the horizontal plane, once the device is correctly orientated thanks a miniature vane. The error in the direction of the flow was one of the main cause of uncertainty. A first objective was, in the case without obstacle, to obtain a steady flow as close as possible to a uniform flow, and then to keep the same parameters in order to quantify the effects of the obstacle. Three possibilities exist in order to adjust the hydraulic characteristics: the discharge of the pumps feeding the channel which may vary from 50 to 300 l/s, the upstream water delivery level, and finally a downstream gate. Adjustment was obtained in the case without obstacle in which the water depth in the main channel remains similar from upstream to downstream. However, it was experimentally difficult to reach a situation in which no flow at all was exchanged between the main channel and the flood plain. This shows the inability of water depth measurements to check the flow uniformity in the case of a compound channel. The velocity field is the second measurement which has to be performed in order to characterize the momentum and mass transfers between the two beds. Only two components are available by using the miniature propeller. This allows to observe the transfers but the mechanisms involved are strongly influenced by 3D structures. Then, measurements were also performed by an ADV probe which shown vertical components of the velocity when a strong conveyance decrease in placed in the flood plain. They can be compared to the one previously observed by LDA around a simple obstacle in a straight rectangular channel.
On décrit une expérience dans un canal du laboratoire de la CNR, qui comprend un lit composé en courbe. Les mesures de hauteur d'eau et de vitesses révèlent un transfert d'un lit à l'autre. Ces transferts sont accentués par la présence d'ouvrages barrant le lit majeur. Une des questions soulevée est l'ajustement de la répartition des débits dans chaque lit à l'amont du modèle.