FTimeXer: Frequency-aware Time-series Transformer with Exogenous variables for Robust Carbon Footprint Forecasting

TL;DR

FTimeXer is a frequency-aware Transformer model that improves power grid carbon footprint forecasting by effectively capturing periodic patterns and handling irregular exogenous inputs, leading to more reliable predictions.

Contribution

The paper introduces FTimeXer, a novel frequency-aware Transformer with a robust training scheme for improved carbon footprint forecasting under challenging conditions.

Findings

FTimeXer outperforms strong baselines on three real-world datasets.

The frequency branch effectively captures multi-scale periodicity.

Robust training enhances stability and reduces spurious correlations.

Abstract

Accurate and up-to-date forecasting of the power grid's carbon footprint is crucial for effective product carbon footprint (PCF) accounting and informed decarbonization decisions. However, the carbon intensity of the grid exhibits high non-stationarity, and existing methods often struggle to effectively leverage periodic and oscillatory patterns. Furthermore, these methods tend to perform poorly when confronted with irregular exogenous inputs, such as missing data or misalignment. To tackle these challenges, we propose FTimeXer, a frequency-aware time-series Transformer designed with a robust training scheme that accommodates exogenous factors. FTimeXer features an Fast Fourier Transform (FFT)-driven frequency branch combined with gated time-frequency fusion, allowing it to capture multi-scale periodicity effectively. It also employs stochastic exogenous masking in conjunction with consistency regularization, which helps reduce spurious correlations and enhance stability. Experiments conducted on three real-world datasets show consistent improvements over strong baselines. As a result, these enhancements lead to more reliable forecasts of grid carbon factors, which are essential for effective PCF accounting and informed decision-making regarding decarbonization.