Visualisation in 3-D can significantly improve the interpretation and understanding of imaged NDT data sets. The advantages of using 3-D graphics to assist in the interpretation of ultrasonic data are discussed in the...
详细信息
Visualisation in 3-D can significantly improve the interpretation and understanding of imaged NDT data sets. The advantages of using 3-D graphics to assist in the interpretation of ultrasonic data are discussed in the context of an advanced software environment for the reconstruction, visualisation and analysis of 3-D images within a component CAD model. The software combines the analysis features of established 2-D packages with facilities for real-time data rotation, interactive orthogonal/oblique slicing and 'true' image reconstruction, where scanning-surface shape and reflection from component boundaries are accounted for through interaction with the full 3-D model of the component. A number of novel facilities exploit the graphics capability of the system. These include the overlay of 3-D images with individual control of image transparency;a floating tooltip window for interrogation of data point co-ordinates and amplitude;image annotation tools, including 3-D distance measurement;and automated defect sizing based on '6dB drop' and,maximum amplitude' methods. A graphical user interface has also been designed for a well established flaw response model, which allows the user to easily specify the flaw size, shape, orientation and location;probe parameters and scan pattern on the component. The output is presented as a simulated ultrasound image.
Two multilayer-perceptron (MLP) artificial-neural-network (ANN) committee methods and a mathematical-macromodelling method for determining the concentrations of gases in a known gas mixture from the outputs of an arra...
详细信息
Two multilayer-perceptron (MLP) artificial-neural-network (ANN) committee methods and a mathematical-macromodelling method for determining the concentrations of gases in a known gas mixture from the outputs of an array of tin-oxide gas sensors placed in the mixtures are described. The committee approach is also used to determine the associated error bars. A large set of artificial training data generated from the small set of experimental data was used to train the MLPs. For the Bayesian-trained committee methods average predicted concentration errors of 1.66 to 9.49% were obtained. The macromodelling method resulted in errors of 19.7 to 33%, but was much easier to implement and faster. Training by back propagation gave much worse accuracy. The average calculated error bars were in good agreement with the actual errors in prediction. The concentration errors were comparable to those yielded by other methods and were at least partly determined by the original errors in the concentration measurements.
In 1954 while reviewing the theory of communication and cybernetics the late Professor Dennis Gabor presented a new mathematical principle for the design of advanced computers. During our work on these computers we fo...
详细信息
In 1954 while reviewing the theory of communication and cybernetics the late Professor Dennis Gabor presented a new mathematical principle for the design of advanced computers. During our work on these computers we found that the Gabor formulation can be further advanced to include more recent developments in Lie algebras and geometric probability, giving rise to a new computing principle.
暂无评论