Simulate to build better
Problem being addressed
Structural design is a process to design the skeleton of a building and ensure its strength, stability, and rigidity under various load conditions. The development of powerful computers and structural simulation programs have made designing complex buildings possible; however, structural design in practice has been a laborious, slow, and iterative process for decades.
An end-to-end learning pipeline to solve the size design optimization problem, which is to design the optimal cross-sections for columns and beams, given the design objectives and building code as constraints. A graph neural network is pre-trained as a surrogate model to not only replace the structural simulation for speed but also use its differentiable nature to provide gradient signals to the other graph neural network for size optimization.
Advantages of this solution
The results show that the pre-trained surrogate model can predict simulation results accurately, and the trained optimization model demonstrates the capability of designing convincing cross-section designs for buildings under various scenarios.
Solution originally applied in these industries
Possible New Application of the Work
Aerospace & Defence Sector
Structural engineering is applied in aircraft and spacecraft construction, which is quite laborious and time-consuming, so simulation in structural design will significantly improve the project efficiency.
Automotive industry can benefit from accurate structural design simulation in order to improve the vehicle safety and its ergonometrics.
Source URL: #############