Artificial intelligence on board dredgers for optimal land reclamation
Type:
Presented during:
CEDA Dredging Days 2007 - The Day After we Stop Dredging, Rotterdam
Authors:
Braaksma J, Babuúka R, Klaassens J B, Delft University of Technology; Osnabrugge J and de Keizer C, IHC Systems, The Netherlands
Abstract: Large land reclamation projects are in general realized via dredging operations. The efficiency of these operations is strongly influenced by the properties of the dredged soil and the skills of the operators. The soil influences start with the necessary force to excavate the soil from the bottom. The excavated soil is transported by means of water via pumps and pipe lines. The behavior of this hydraulic transport process is mainly determined by the grain size of the soil, the dredging depth and the velocity and density of the mixture. But also the sedimentation speed in a hopper is strongly dependent on these parameters. An experienced operator observes changes in behavior caused by changing soil properties and the dredging depth and adapts his control actions to maintain optimal process control. A serious problem nowadays is that too few experienced operators are available. Therefore automatic control techniques can give a solution.
In this paper it is shown that the required adaptive behavior can be realized automatically by using artificial intelligence and advanced control techniques. The principle is to observe the changing behavior - as an experienced operator does - and to translate this behavior into understandable soil parameters such as grain sizes and sedimentation velocity. With this knowledge the dredge operation can be optimized.
This paper gives an overview of what has been realized in this field during the last years, especially for the loading process of a trailing suction hopper dredgers and the discharge process via long discharge pipelines (up to 15 km) for both cutter suction dredgers and trailing suction hopper dredgers. Figures are presented of the expected and realized performance improvements both in comparison with experienced and less experienced
operators.
Keywords: dredging, artificial intelligence, advanced control, optimization, hydraulic transport