Robust Data-Driven Error Compensation for a Battery Model
This work aims to improve the accuracy and robustness of battery models for automotive drivetrain development by leveraging large, real-world battery datasets, which is an incremental improvement for engineers.
The paper addresses the underutilization of large battery datasets for improving battery model accuracy due to non-uniform excitation. It introduces a data-driven error model, specifically a neural network, to compensate for dynamic errors in an existing physical model, demonstrating similar improvement and increased robustness across five datasets when gradually limiting compensation outside data boundaries.
- This work has been submitted to IFAC for possible publication - Models of traction batteries are an essential tool throughout the development of automotive drivetrains. Surprisingly, today's massively collected battery data is not yet used for more accurate and reliable simulations. Primarily, the non-uniform excitation during regular battery operations prevent a consequent utilization of such measurements. Hence, there is a need for methods which enable robust models based on large datasets. For that reason, a data-driven error model is introduced enhancing an existing physically motivated model. A neural network compensates the existing dynamic error and is further limited based on a description of the underlying data. This paper tries to verify the effectiveness and robustness of the general setup and additionally evaluates a one-class support vector machine as the proposed model for the training data distribution. Based on a five datasets it is shown, that gradually limiting the data-driven error compensation outside the boundary leads to a similar improvement and an increased overall robustness.