A new FEA approach utilizing the concept of visual computing is proposed. Conventional FEA is improved by a new visual computing method. This new method is based on rapid data interchange among program modules through the use of shared computer memory. The concept enables easy control of the analysis process through interactive user input. It also provides the possibility for on-line visualization and animation in real time of various one-dimensional scalar quantities and one-dimensional and two-dimensional distributions, using a fast computer graphics engine. By visualizing the numerical data in progress, the user can easily detect errors in the initial data, achieving rapid and accurate analysis with reduced computation cost. The proposed approach is also very useful for analysis of time-periodical, nonlinear or any other problems having many successive iterations and solutions.
展开▼
