The industry's recognized authority for design, engineering and application of equipment and services in the global ocean community
Issue link: http://sea-technology.epubxp.com/i/653412
www.sea-technology.com March 2016 / st 11 producing high data density, and the points were oversam- pled up to 10 times, which increases the reliability of the data sets. Initial Results Right after the mechanical installation of the system, the frst baseline survey was performed. The system was con- fgured to monitor the area with a radius of around 10 m in front of the dolphin. The processing algorithm was confg- ured to generate a grid with cell size of 0.2 m, and the alarm threshold was set to 0.4 m calculated as vertical deviation on the seabed. The data from the baseline survey were used as reference data for calculation of deviations on the sea- bed. The data analysis for the period of 10 months showed only small changes on the seabed. The frst deviations oc- curred after three months monitoring the seabed, on the far right edge of the data set, approximately 10 m away from the foundation of the dolphin. The change in the seabed could not be characterized as scour but only as small ero- sion, produced probably as a result of the vertices formed by the thrusters of the vessels moving in that area. The vertical change on the seabed was around 0.4 m, which was the exact alarm threshold value set in the system confguration. After the threshold value was exceeded, the system execut- ed a scour alarm. Due to the terrain confguration and big- ger slope on the erosion area, shadow zones are detected. Due to the small changes on the seabed during the frst six months of operation, and for the sake of testing the ca- pabilities of the K-Observer system and the DAS sonar, a simulation test was performed on July 7, 2015 using fve small sandbags and a reference sphere. The sandbags were around 20 to 25 cm in diameter, and the reference sphere was 40 cm in diameter. The objects were deployed on the seabed in the area of interest and were supposed to simulate changes on the seabed. In the next stage, a few new scans were performed and the data were processed automatically with the K-Observer system and compared with the scans from the previous days. The results showed that even though there is a small layer of mud on the seabed and all objects (sandbags and refer- ence sphere) sunk up to some extent in the mud, they could be clearly identifed on the point cloud and 3D grid data. Generally, the data generated by the DAS sonar were dense enough to detect objects smaller than 20 cm, which is an excellent result. Moreover, compared to previous scans when the ob- jects were not deployed on the seabed, the difference grid showed a deviation of around 0.35 m on the position of the reference sphere and 0.2-m deviation on the position of the sandbags. Consequently, the system automatically showed an alarm, in this case sediment buildup (positive scour). Calculating the total volume difference between the same measurements for the complete scanned area, there is positive change only in the areas where the objects were deployed. The volume changes caused by the simulation are the same as the erosion caused by the natural forces. The grid cell size was 0.3 cu. m, the positive volume difference was 1.4 cu. m, the negative volume difference was -1.4 cu. m, and the highest vertical deviation was 0.35 m. Finally, the objects could also be detected on the cross- section graphs using data sets for comparison from before (Top) Baseline survey of the Port of Hamburg. (Bottom) Cross-section graph using data sets for compari- son from the beginning of the in- stallation and after three months.