Efficient mid-term forecasting of hourly electricity load using generalized additive models

2024-05-27

arXiv.org

AI Summary

Accurate mid-to-long-term (weeks to one year) hourly electricity load forecasting remains challenging due to complex interactions among multi-scale seasonality, nonlinear weather dependence, socioeconomic nonstationarity, and autoregressive dynamics. Method: This paper proposes a high-accuracy, fully interpretable Generalized Additive Model (GAM) featuring: (i) interpretable P-spline smoothing; (ii) ETS-driven persistent modeling of socioeconomic states; (iii) hierarchical holiday representation across fixed dates, weekdays, and vacation seasons; (iv) nonlinear temperature response modeling; and (v) autoregressive residual correction. Contribution/Results: The framework jointly captures seasonal, meteorological, socioeconomic, and dynamic effects in a unified, transparent manner. Evaluated on over nine years of hourly data from 24 European countries, it significantly outperforms existing state-of-the-art methods—achieving accuracy comparable to transmission system operators’ day-ahead forecasts—while enabling sub-second inference and scalable, industrial-grade real-time deployment for multi-year hourly forecasting.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Machine Learning: Mixture of Experts (MoE)
• Intelligent Robots: State Estimation

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Accurate mid-term (weeks to one year) hourly electricity load forecasts are essential for strategic decision-making in power plant operation, ensuring supply security and grid stability, planning and building energy storage systems, and energy trading. While numerous models effectively predict short-term (hours to a few days) hourly load, mid-term forecasting solutions remain scarce. In mid-term load forecasting, capturing the multifaceted characteristics of load, including daily, weekly and annual seasonal patterns, as well as autoregressive effects, weather and holiday impacts, and socio-economic non-stationarities, presents significant modeling challenges. To address these challenges, we propose a novel forecasting method using Generalized Additive Models (GAMs) built from interpretable P-splines that is enhanced with autoregressive post-processing. This model incorporates smoothed temperatures, Error-Trend-Seasonal (ETS) modeled and persistently forecasted non-stationary socio-economic states, a nuanced representation of effects from vacation periods, fixed date and weekday holidays, and seasonal information as inputs. The proposed model is evaluated using load data from 24 European countries over more than 9 years (2015-2024). This analysis demonstrates that the model not only has significantly enhanced forecasting accuracy compared to state-of-the-art methods but also offers valuable insights into the influence of individual components on predicted load, given its full interpretability. Achieving performance akin to day-ahead Transmission System Operator (TSO) forecasts, with computation times of just a few seconds for several years of hourly data, underscores the potential of the model for practical application in the power system industry.

Problem

• Mid-term hourly electricity load forecasting
• Capturing multifaceted load characteristics
• Enhancing forecasting accuracy and interpretability

Innovation

• Generalized Additive Models (GAMs)
• Autoregressive post-processing
• Smoothed temperatures and ETS modeling