Shape Optimization of Device Components

Automated testing to optimize material usage through shaped optimization.

Solution by: Data Monsters

Date: March 2021

Challenge

When designing various physical devices, it’s necessary to minimize the cost of materials while still meeting strength requirements. For this, destructive testing of digital 3D models of device components in physical simulation systems is used. These experiments use manually-selected parameters of the components, and are time-consuming. The challenge is to create a tool which would automate this routine.

Solution

A pipeline was developed, including processing the CAD file, which automatically runs experiments in a physical simulation system, and optimizes the component parameters using a Bayesian optimization tool. Since the parts of the pipeline are located on different servers, data exchange is carried out using a secure protocol. The user sets experimental parameters such as boundary conditions and cell size, and the system calculates the optimal shape of the device components, optimizing material consumption while meeting strength requirements.

Technologies used

• Finite element analysis
• Bayesian optimization
• Remote procedure call