Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Probabilistic Surrogate Model for Accelerating the Design of Electric Vehicle Battery Enclosures for Crash Performance

Shadab Anwar Shaikh, Harish Cherukuri, Kranthi Balusu, Ram Devanathan, Ayoub Soulami

TL;DR

This work tackles accelerating the design of CFRP battery enclosures for EV crashworthiness by building a probabilistic surrogate via Gaussian Process Regression from high-throughput FE data. The GP model provides accurate predictions of crash metrics $\{F_p,\ \text{CLE},\ \text{SEA},\ \Delta Y_{node}\}$ with uncertainty estimates, dramatically reducing computation time from hours to milliseconds per prediction. Key contributions include an end-to-end data-generation pipeline (thermoforming+crash), ARD-enabled Matérn 3/2 GPR that outperforms linear baselines, and a Monte Carlo uncertainty-propagation analysis showing limited sensitivity of outputs to input variations. This approach enables rapid, uncertainty-aware design optimization of composite battery enclosures and can be extended to additional crash modes and material-parameter variations.

Abstract

This paper presents a probabilistic surrogate model for the accelerated design of electric vehicle battery enclosures with a focus on crash performance. The study integrates high-throughput finite element simulations and Gaussian Process Regression to develop a surrogate model that predicts crash parameters with high accuracy while providing uncertainty estimates. The model was trained using data generated from thermoforming and crash simulations over a range of material and process parameters. Validation against new simulation data demonstrated the model's predictive accuracy with mean absolute percentage errors within 8.08% for all output variables. Additionally, a Monte Carlo uncertainty propagation study revealed the impact of input variability on outputs. The results highlight the efficacy of the Gaussian Process Regression model in capturing complex relationships within the dataset, offering a robust and efficient tool for the design optimization of composite battery enclosures.

Probabilistic Surrogate Model for Accelerating the Design of Electric Vehicle Battery Enclosures for Crash Performance

TL;DR

This work tackles accelerating the design of CFRP battery enclosures for EV crashworthiness by building a probabilistic surrogate via Gaussian Process Regression from high-throughput FE data. The GP model provides accurate predictions of crash metrics with uncertainty estimates, dramatically reducing computation time from hours to milliseconds per prediction. Key contributions include an end-to-end data-generation pipeline (thermoforming+crash), ARD-enabled Matérn 3/2 GPR that outperforms linear baselines, and a Monte Carlo uncertainty-propagation analysis showing limited sensitivity of outputs to input variations. This approach enables rapid, uncertainty-aware design optimization of composite battery enclosures and can be extended to additional crash modes and material-parameter variations.

Abstract

This paper presents a probabilistic surrogate model for the accelerated design of electric vehicle battery enclosures with a focus on crash performance. The study integrates high-throughput finite element simulations and Gaussian Process Regression to develop a surrogate model that predicts crash parameters with high accuracy while providing uncertainty estimates. The model was trained using data generated from thermoforming and crash simulations over a range of material and process parameters. Validation against new simulation data demonstrated the model's predictive accuracy with mean absolute percentage errors within 8.08% for all output variables. Additionally, a Monte Carlo uncertainty propagation study revealed the impact of input variability on outputs. The results highlight the efficacy of the Gaussian Process Regression model in capturing complex relationships within the dataset, offering a robust and efficient tool for the design optimization of composite battery enclosures.
Paper Structure (21 sections, 12 equations, 10 figures, 16 tables)

This paper contains 21 sections, 12 equations, 10 figures, 16 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Current simulation workflow for enclosure design
  4. Surrogate Modeling using ML
  5. Gaussian Process Regression
  6. Methodology
  7. Data generation for surrogate model
  8. Geometry of the battery enclosure
  9. Thermoforming simulation setup
  10. Crash simulation setup
  11. Automation pipeline
  12. Post-processing
  13. Results and Discussion
  14. Predictions on the holdout set
  15. Repeated K-Fold Cross Validation
  16. ...and 6 more sections

Figures (10)

  • Figure 1: Schematic of the current simulation workflow for enclosure design shaikh2023finite
  • Figure 2: Summary of overall methodology
  • Figure 3: Geometry of the battery enclosure used in the analysis kulkarni2023investigation
  • Figure 4: Schematic for thermoforming and crash simulation kulkarni2023investigation
  • Figure 5: Schematic of a automation pipeline
  • ...and 5 more figures