Application of Scientific Data Visualization Techniques
Helge Langseth · NORA - Norwegian Open Research Archives · 2002
This document is a thesis for the master of science degree at the University of Oslo. Working with the Composite Hull Embedded Sensor System project group, I have worked with visualizing data from a ship hull monitoring system developed at the Norwegian Defense Research Establishment. The instrumentation Among several other vessels, the CHESS project group has instrumented the surface effect ship fast patrol boat HNoMS Skjold of the Norwegian navy. This ship is a pre-series vessel, and the findings from the ship hull monitoring system has been part of the testing and design verification of the vessel. This ship has been instrumented with 56 fiber Bragg grating sensors multiplexed on multiple fibers and interrogated by Fabry-Perot scanning filters and unbalanced fiber interferometers. In addition, a wave radar measures the distance from the bow to the water level below, a GPS unit measures exact position and a motion reference unit measures ship rotations in all three directions. Data transfer and storage Data from the sensors mentioned above are sampled by several personal computers connected together in a computer network. These PCs have synchronized internal clocks and communicate over the network using the TCP/IP protocol and sockets communication. A software library with common functions for data structures and communication has been developed for the project and currently offers modules for data collection, transmission and storage, signal processing and filtering, simple visualization, and so on. For storage and post-processing, data sets have been stored as Matlab data files and have been processed in Matlab. This has made it easier for the members of the CHESS project group to experiment with the data and look for correlations between movements and strain induced on the hull of the ship. For this reason, the visualization program described in this thesis has been designed to read Matlab data files with the CHESS project's data structure. Visualization From the Matlab files or through TCP/IP sockets, huge amounts of data are available. To better use these data, some sort of visualization is necessary, and such a module has been the subject of this thesis. The program designed and partly implemented draws a real-time animation of the ship as it heaves, rolls and stamps through the ocean in three dimentions. Additionally, it will deform along the horizontal XY- and XZ-plane in bending modes and torsion and water will be drawn underneath it. In a separate window, the user has access to plotted graph data from ship sensors and a two dimentional plot of the meeting wave profile. Development is done on an ordinary computer running the Linux operating system. The current version is 7.2 of the Red Hat distribution. On this PC, programming is done in OpenGL and C++ using the Qt C++ toolkit by Trolltech AS, a toolkit also used to make the K Desktop Environment (KDE) for Linux.