Simplifying Complex Structures: A Faster Way to Predict Vibrations

What is it about?

Engineers who work with massive structures, such as skyscrapers or intricate bridges, need to predict how these structures will react to forces like earthquakes. This prediction requires complex calculations, which can take a lot of time. Reduced order modeling (ROM) is a technique that simplifies the intricate model into a smaller, more manageable one that captures the key vibrations, such as the initial swaying motions. This enables faster calculations, and the accuracy of the results is not compromised. One existing technique, called IOR-substructuring, is slow and sometimes produces inaccurate results due to instability issues. However, this paper has introduced a modified IOR-substructuring method that addresses these problems.

Featured Image

Photo by NASA on Unsplash

Photo by NASA on Unsplash

Why is it important?

The proposed method employs interface boundary reduction to speed up the calculations significantly, and it avoids instability issues, ensuring reliable results. We compared this new method with other ROM techniques and found that it delivers high accuracy, matches the accuracy of the best existing methods, and is considerably faster than the original IOR-substructuring method. Additionally, it requires less computer memory, making it more memory-efficient. In summary, the new method provides a faster and more reliable way for engineers to analyze the vibrations of complex structures.