Deep Learning-Based Analysis of Power Consumption in Gasoline, Electric, and Hybrid Vehicles

TL;DR

This paper presents a scalable deep learning approach for accurately predicting power consumption in various vehicle types, improving efficiency analysis without relying on physical models or specialized instruments.

Contribution

Introduces a data-driven method combining traditional machine learning and deep neural networks for power consumption estimation across multiple vehicle platforms.

Findings

High accuracy in ICE power prediction with errors under 0.3%.

Transformer and LSTM models outperform others for EV and HEV predictions.

Greater variability in EV and HEV datasets highlights complexity in power management.

Abstract

Accurate power consumption prediction is crucial for improving efficiency and reducing environmental impact, yet traditional methods relying on specialized instruments or rigid physical models are impractical for large-scale, real-world deployment. This study introduces a scalable data-driven method using powertrain dynamic feature sets and both traditional machine learning and deep neural networks to estimate instantaneous and cumulative power consumption in internal combustion engine (ICE), electric vehicle (EV), and hybrid electric vehicle (HEV) platforms. ICE models achieved high instantaneous accuracy with mean absolute error and root mean squared error on the order of $10^{-3}$, and cumulative errors under 3%. Transformer and long short-term memory models performed best for EVs and HEVs, with cumulative errors below 4.1% and 2.1%, respectively. Results confirm the approach's effectiveness across vehicles and models. Uncertainty analysis revealed greater variability in EV and HEV datasets than ICE, due to complex power management, emphasizing the need for robust models for advanced powertrains.