EnergyPatchTST: Multi-scale Time Series Transformers with Uncertainty Estimation for Energy Forecasting

TL;DR

EnergyPatchTST is a novel multi-scale transformer model for energy time series forecasting that incorporates uncertainty estimation, future variable integration, and pre-training to improve accuracy and reliability over existing methods.

Contribution

The paper introduces EnergyPatchTST, a transformer-based model with multi-scale feature extraction, uncertainty estimation, and pre-training tailored for energy forecasting.

Findings

Prediction error reduced by 7-12% compared to other methods.

Provides reliable uncertainty estimates for energy forecasts.

Effective in limited data scenarios through pre-training.

Abstract

Accurate and reliable energy time series prediction is of great significance for power generation planning and allocation. At present, deep learning time series prediction has become the mainstream method. However, the multi-scale time dynamics and the irregularity of real data lead to the limitations of the existing methods. Therefore, we propose EnergyPatchTST, which is an extension of the Patch Time Series Transformer specially designed for energy forecasting. The main innovations of our method are as follows: (1) multi-scale feature extraction mechanism to capture patterns with different time resolutions; (2) probability prediction framework to estimate uncertainty through Monte Carlo elimination; (3) integration path of future known variables (such as temperature and wind conditions); And (4) Pre-training and Fine-tuning examples to enhance the performance of limited energy data sets. A series of experiments on common energy data sets show that EnergyPatchTST is superior to other commonly used methods, the prediction error is reduced by 7-12%, and reliable uncertainty estimation is provided, which provides an important reference for time series prediction in the energy field.