Translation waves induced by high head lock systems

Abstract

The main goal in this research is to predict how translation waves evolve in a canal and what the consequences for the workability of water-borne construction equipment are. Translation waves arise due to filling and emptying of lock systems. In this research the measures to reduce effects of translation waves are control options related to the lock complex only. Adaptations to the water-borne construction equipment or to the bathymetry of the canal are excluded from this research. Currents induced by other external factors than translation waves induced by the lock complex, are not investigated. The research is applied to a case study, the Juliana Canal. Construction works are performed to make the canal attainable for vessels of CEMT-class Vb. To predict the adaptations of water level changes and currents due to translation waves a 1D-model is developed. In the model, theory from Thijsse (Thijsse, 1935) on the development of a translation wave is applied and extended on more complex discharge extractions from the canal. The propagating and reflecting waves through the canal are calculated. The bathymetry of the canal-system determines the transition points and reflections. To obtain a good representation of the translation wave in this canal, a set of 42 reflections were included in this model. If the currents in the canal become too large, unworkable construction situations arise resulting in downtime. Two control options can be done at the locks. The first option is regulating the magnitude of the discharge to fill the lock chamber. This is done by defining the number of culverts used during filling. Four scenarios are investigated, with: 2, 1 or 4 culverts used and with 2 chambers filled simultaneously. The second option is to vary the time between sequent fillings. Six scenarios for this option are investigated, with: 5, 10 and 15 minutes between subsequent waves, chambers filled with 1 or 2 culverts. None of the investigated scenarios result in no downtime but the downtime can be reduced significantly. The most favourable scenario from the first group of control options, vary between 1, 2 and 4 culverts, depending on the location. Simultaneous filling should always be avoided. In the second group of control options, the scenario where the lock chamber is filled with one culvert and 15 minutes between sequent waves, results at 4 of the 5 locations in the least downtime. At the other location the best option is the scenario where the chamber is filled with 2 culverts and also 15 minutes between the waves. There is not a scenario that is the best scenario for all locations along the canal. The resulting water level change and current are the summation of the original wave and reflection waves. The reflection waves are created by e.g. reflections at a port or changes in the width of the canal. The distance of the construction location to these transition points determine which reflections, and with what magnitude, are at that specific location. This makes the net water level change and net current very location dependent. For each location the model has to be run with different scenarios. Concluding, the model gives a good representation of the situation in reality, for the modelled locations in canal. For future projects, the model can relatively easy be adapted to another canal-system by changing the bathymetry, lock discharges and transition points. The model is applicable to gain insights of a system for designs. It can be used as a decision tool during construction works as well due to the short calculation period. The use of this model can forecast the working conditions and limit the downtime.