Short-term Load Forecasting Based on Hybrid Strategy Using Warm-start Gradient Tree Boosting

TL;DR

This paper introduces a novel hybrid ensemble method called Warm-start Gradient Tree Boosting (WGTB) for short-term load forecasting, combining diverse inference models to improve accuracy by balancing bias and variance.

Contribution

The paper proposes WGTB, a new ensemble strategy that effectively combines different inference models using warm-start, bagging, and boosting techniques for enhanced load forecasting.

Findings

WGTB outperforms individual models on real datasets.

Hybrid ensemble reduces bias and variance effectively.

Validated on datasets from State Grid Corporation of China.

Abstract

A deep-learning-based hybrid strategy for short-term load forecasting is presented. The strategy proposes a novel tree-based ensemble method Warm-start Gradient Tree Boosting (WGTB). Current strategies either ensemble submodels of a single type, which fail to take advantage of the statistical strengths of different inference models. Or they simply sum the outputs from completely different inference models, which doesn't maximize the potential of ensemble. Inspired by the bias-variance trade-off, WGTB is proposed and tailored to the great disparity among different inference models on accuracy, volatility and linearity. The complete strategy integrates four different inference models of different capacities. WGTB then ensembles their outputs by a warm-start and a hybrid of bagging and boosting, which lowers bias and variance concurrently. It is validated on two real datasets from State Grid Corporation of China of hourly resolution. The result demonstrates the effectiveness of the proposed strategy that hybridizes the statistical strengths of both low-bias and low-variance inference models.