Data Model Design for Explainable Machine Learning-based Electricity Applications

TL;DR

This paper introduces a taxonomy for structuring multivariate energy data to improve machine learning models in smart grid applications, validated through household electricity forecasting experiments.

Contribution

It proposes a novel taxonomy for energy data types that guides data model development and enhances model interpretability in electricity forecasting.

Findings

The taxonomy effectively guides feature selection for different models.

Including domain, contextual, and behavioral features improves forecasting accuracy.

Feature importance analysis explains individual feature contributions.

Abstract

The transition from traditional power grids to smart grids, significant increase in the use of renewable energy sources, and soaring electricity prices has triggered a digital transformation of the energy infrastructure that enables new, data driven, applications often supported by machine learning models. However, the majority of the developed machine learning models rely on univariate data. To date, a structured study considering the role meta-data and additional measurements resulting in multivariate data is missing. In this paper we propose a taxonomy that identifies and structures various types of data related to energy applications. The taxonomy can be used to guide application specific data model development for training machine learning models. Focusing on a household electricity forecasting application, we validate the effectiveness of the proposed taxonomy in guiding the selection of the features for various types of models. As such, we study of the effect of domain, contextual and behavioral features on the forecasting accuracy of four interpretable machine learning techniques and three openly available datasets. Finally, using a feature importance techniques, we explain individual feature contributions to the forecasting accuracy.