Automated feature engineering improves energy consumption forecasting
Automated feature engineering, AutoML, and decision-focused learning for improved energy consumption forecasting
Artificial Intelligence
Summary
Predicting how much energy people and businesses will use is important but often needs experts to choose the right features for machine learning models. The authors created an automated method called AutoEnergy that designs useful features from data automatically, making predictions more accurate and faster across many real-world energy datasets. They also combined this method with decision-focused learning to better manage battery storage, saving significant operating costs in a UK property case. This shows that automating feature design can reduce manual work, improve forecasts, and help save money in energy management.
What this means in practice
- •For energy management teams: Improve energy demand forecasting by automatically generating and selecting features to enhance planning accuracy across diverse energy sectors.
- •For battery storage system operators: Optimize charge and discharge decisions by integrating automated forecasting features into decision-focused learning models to cut operating costs.
Authors
Nasser Alkhulaifi
Abstract
The rising cost and demand for energy, together with environmental sustainability goals, create major challenges for energy management. Energy Consumption Forecasting (ECF) supports planning by predicting future consumption, but Machine Learning (ML) models for ECF often depend on expert-driven Feature Engineering (FE). This thesis addresses that dependence through three contributions. First, it establishes and evaluates a comprehensive FE pipeline for ECF and investigates domain-specific features. Second, it introduces AutoEnergy, a domain-tailored automated FE algorithm that generates interpretable features from timestamps and lagged consumption and integrates with AutoML for end-to-end ECF modelling. Across eighteen real-world energy datasets spanning residential, commercial, industrial, renewable, and grid domains, AutoEnergy reduces forecasting error by 19.52%-84.72% relative to baseline AutoML and established automated FE methods, while running 1.31-4.41 times faster, with gains varying by dataset. Third, AutoEnergy is integrated with Decision-Focused Learning (DFL) for a Battery Energy Storage System problem, jointly forecasting electricity prices and demand while optimising charging and discharging decisions. On a real-world UK property dataset, this approach reduces operating costs by 22.9%-56.5% compared with the same DFL models without automated FE. Overall, the results show that domain-specific automated FE can reduce reliance on manual feature design, improve forecasting accuracy, and translate predictive gains into measurable operational benefits in energy management.