Multi-Model Analysis of O-Ring Compression and Relaxation Force Predictions Using Machine Learning and CAE Simulations

O-rings play a critical role in ensuring leak-proof seals in a wide range of engineering systems. Accurate prediction of their compression and relaxation behavior under various material and geometric configurations is essential for optimal design and reliability. This study presents an analysis of machine learning techniques to predict two key performance outputs, compression force and relaxation force (after 10 minutes) trained on computer-aided engineering (CAE) simulation data. The experimental setup was represented in CAE simulation and the results were compared with experimental data conducted at ZF test facilities. Simulation results correlated well with the experimental data (deviation was less than the 5%). To create a dataset for training machine learning (ML) models, realistic ranges for the input parameters such as hardness and geometrical parameters were determined, and simulation data were generated using design of experiments (DOE). Multiple ML models were developed and evaluated based on error metrics and diagnostic plots. The objective of this study was to identify the most effective algorithm for capturing complex nonlinear relationships between input parameters and target performance metrics. Neural Network (NN) model performed well in predicting the forces in the considered design space. For new data outside the training set, it was able to predict the forces with minimal error (mean absolute percentage error of 2.5% and R² of 0.99). A tool was created with this ML model, which helps designers to take a data informed decision and enhance product development process. Using this tool forces can now be obtained instantly, whereas the complete process (design and simulation) requires significant time. This approach shows the feasibility of combining machine learning with engineering simulations to speed up the design cycle and lower development costs.