Model-based analysis to identify the impact of factors affecting electricity gaps during COVID-19: A case study in Germany

Nanae Kaneko, Yu Fujimoto, Hans‐Arno Jacobsen, Yasuhiro Hayashi

TL;DR

The COVID-19 pandemic drastically altered electricity demand, creating a "demand gap." This paper introduces a novel methodology combining time-series forecasting (ARIMAX) and sparse modeling (PLAMs) to pinpoint the specific economic and lifestyle factors driving this gap in Germany. The approach effectively identifies key variables and quantifies their impact, offering crucial insights for energy sector stakeholders navigating unprecedented disruptions.

What Problem Does This Paper Address?

The COVID-19 pandemic, with its lockdowns and shifts in economic and social behavior, significantly disrupted electricity consumption patterns. This led to a widening "demand gap" – the difference between actual electricity usage and what would have been predicted based on pre-pandemic trends. While many studies have touched on the impact of lockdowns, the specific influence of broader economic shifts and evolving lifestyle choices on this demand gap remained largely unexplored. Understanding these drivers is critical for electricity utilities and policymakers to develop effective strategies for grid management, infrastructure investment, and energy conservation, especially in the face of future crises.

Methods

The authors propose a three-step, data-driven methodology:

  1. Situation-Dependent Modeling (Sparse PLAMs): They first construct "partially linear additive models" (PLAMs) that can capture both linear and nonlinear relationships between numerous potential explanatory variables and hourly electricity demand. A key innovation here is the use of sparse enumeration to select a parsimonious set of the most critical variables for each specific time period (month/hour), making the models more interpretable and efficient.
  2. Hypothetical Non-COVID-19 Scenario Estimation (ARIMAX): To establish a baseline, they use Autoregressive Integrated Moving Average with an exogenous variable (ARIMAX) models. These models are trained on pre-pandemic data to forecast what electricity demand and key influencing variables would have been in the absence of COVID-19.
  3. Demand Gap Analysis and Contribution Evaluation: By comparing the actual COVID-19 demand scenario (using observed data) with the estimated non-COVID-19 scenario, they calculate the demand gap. Crucially, they then evaluate the additive contribution of each identified key variable to this gap, distinguishing between "deviance-oriented" gaps (caused by variables behaving unexpectedly) and "expected" gaps (where variables fluctuate within normal ranges).

Datasets

The study utilizes hourly electricity usage data for Germany spanning from January 2015 to December 2021. This dataset includes 179 potential explanatory variables categorized into weather, interest rates, stock prices, calendars, and GDP data, sourced from publicly available data.

Key Insights and Ideas

  • Beyond Lockdowns: The pandemic's impact on electricity demand is not solely due to movement restrictions. Economic shifts and evolving consumer habits (like remote work, changes in consumption of goods and services) play a significant, quantifiable role.
  • Seasonality Matters: The influence of key variables on the demand gap varies significantly across seasons and even time of day, highlighting the need for granular, situation-dependent analysis.
  • Sparse Modeling is Key: Identifying a small, critical set of variables using sparse PLAMs is crucial for understanding complex demand dynamics, rather than relying on a large, potentially noisy set of factors.
  • Deviance vs. Expectation: Differentiating between demand gaps caused by unexpected variable behavior (deviance-oriented) and those within expected fluctuations is vital for accurate interpretation and forecasting.

Key Results

  • The proposed methodology effectively identifies key variables impacting the electricity demand gap, demonstrating higher descriptive accuracy than simpler linear or additive models.
  • Analysis of German data revealed that factors like the production index of sports/outdoor equipment and consumer interest in "solar energy" (via Google Trends) showed a negative correlation with electricity demand, suggesting shifts towards outdoor activities and potentially increased adoption of personal solar power.
  • The study identified specific variables, such as the manufacturing of food products, passenger transport by road, and consumer prices for refuse collection, that significantly contributed to the demand gap, with their impact varying by month and hour.
  • The research successfully quantifies the additive contributions of these variables to both the expected and deviance-oriented components of the demand gap.

Limitations and Future Work

  • Post-2021 Impact: The study primarily focuses on 2020-2021. Future work should examine the ongoing effects of the pandemic on electricity demand beyond this period, as consumption patterns continue to evolve.
  • Hierarchical Factor Structure: While the study identifies key variables, exploring a hierarchical structure for factors that indirectly influence the demand gap could offer a more comprehensive understanding.
  • Uncertainty Quantification: While the study touches on confidence intervals, further research could focus on developing forecasting methodologies that more robustly account for the high volatility and uncertainty associated with key variables.
  • ESG Factors: Investigating the relationship between long-term electricity demand and ESG (environmental, social, and governance) factors could provide a deeper physical interpretation of demand dynamics.

Summary updated

Abstract

The recent COVID-19 pandemic has precipitated drastic changes in economic and lifestyle conditions, significantly altering residual electricity demand behavior. This alteration has expanded the demand gap between actual and forecasted electricity usage based on pre-pandemic data, highlighting a critical global issue. Many studies in the pandemic have explored the features of this widening gap, which is impacted by major social events like fast virus spread and lockdowns. However, the influence of factors like economic shifts and lifestyle changes on this demand remains largely unexplored, primarily due to the pandemic's significant effects in these areas. Understanding the essential factors affecting the demand gap is crucial for stakeholders in the electricity sector to develop effective strategies. This study examines the hourly electricity consumption and related factors during the specified period. We present a method combining time-series forecasting and sparse modeling. This helps identify critical factors affecting the electricity demand gap during the pandemic, highlighting the most crucial variables. Utilizing this method, we identify the variables that have undergone significant changes during the pandemic and evaluate their effects on the electricity demand gap. The effectiveness is proven by applying it to the dataset collected in German.
BibTeX
@inproceedings{kaneko2024model,
  author = {Nanae Kaneko and Yu Fujimoto and Hans‐Arno Jacobsen and Yasuhiro Hayashi},
  title = {{Model-based analysis to identify the impact of factors affecting electricity gaps during COVID-19: A case study in Germany}},
  booktitle = {Heliyon},
  year = {2024},
  doi = {10.1016/j.heliyon.2024.e33943},
}
Figure