Back to AI Research

AI Research

Analysing drivers and interdependencies in European... | AI Research

Key Takeaways

  • Analysing drivers and interdependencies in European electricity markets using XAI This research investigates the complex factors that determine electricity p...
  • Electricity markets are inherently complex systems characterised by strong nonlinearities, high-dimensional interactions, and increasing interdependence across regions.
  • This paper addresses this gap by combining DNN models with explainable artificial intelligence (XAI) techniques to analyse the determinants of electricity prices across 39 European bidding zones.
  • We employ SHAP (SHapley Additive exPlanations) to quantify feature contributions and apply and extend SSHAP, an aggregation framework to improve interpretability in high-dimensional settings.
  • The analysis identifies that renewable energy sources, particularly solar, play a disproportionately important role in price formation despite their lower share in total power generation.
Paper AbstractExpand

Electricity markets are inherently complex systems characterised by strong nonlinearities, high-dimensional interactions, and increasing interdependence across regions. While deep neural networks (DNNs) have demonstrated strong predictive capabilities for electricity prices, their lack of interpretability limits their usefulness for understanding the underlying drivers of price formation. This paper addresses this gap by combining DNN models with explainable artificial intelligence (XAI) techniques to analyse the determinants of electricity prices across 39 European bidding zones. We employ SHAP (SHapley Additive exPlanations) to quantify feature contributions and apply and extend SSHAP, an aggregation framework to improve interpretability in high-dimensional settings. The analysis identifies that renewable energy sources, particularly solar, play a disproportionately important role in price formation despite their lower share in total power generation. Gas prices remain a dominant and consistent driver across electricity markets, while interconnections significantly shape price dynamics, highlighting the strong interdependence of European electricity systems. In addition, a synthetic EU-wide electricity market is constructed to explore the counterfactual scenario of a fully integrated market with a single price.

Analysing drivers and interdependencies in European electricity markets using XAI
This research investigates the complex factors that determine electricity prices across 39 European bidding zones. While deep neural networks (DNNs) are excellent at predicting electricity prices, they typically function as "black boxes," making it difficult for policymakers and market participants to understand why prices change. By applying Explainable Artificial Intelligence (XAI) techniques, this study uncovers the underlying drivers of price formation, such as the influence of renewable energy, gas prices, and cross-border interconnections.

How the approach works

The researchers built deep learning models to predict electricity prices using data from the ENTSO-E Transparency Platform, covering the years 2023 to 2025. To make these models interpretable, they employed SHAP (SHapley Additive exPlanations), a method that assigns a contribution value to each input feature. Because the electricity market involves a high volume of interconnected data, the authors utilized an advanced framework called SSHAP. This allows them to group individual data points into "super-features"—such as specific power generation types or neighboring countries—to better understand how these categories influence price movements without the noise of high-dimensional data.

Key findings on price drivers

The analysis reveals that renewable energy sources, particularly solar power, have a disproportionately large impact on price formation. Even though solar accounts for a smaller share of total electricity generation compared to other sources, it plays a leading role in setting price trends. Gas prices remain a consistent and dominant driver across the continent, confirming the market's reliance on fossil fuels. Furthermore, the study highlights the deep interdependence of the European grid; for example, the electricity generation and load data from neighboring countries are often the most significant factors in determining a specific country’s domestic price.

The "What If" scenario

To explore the future of the European energy system, the researchers constructed a synthetic, EU-wide electricity market model. This counterfactual scenario simulates what would happen if the European market were fully integrated with a single clearing price. By using this model, the authors were able to quantify the degree of market integration and demonstrate how interconnectedness shapes price dynamics. This provides a clearer picture of how a unified European energy policy might behave compared to the current fragmented system of 39 distinct bidding zones.

Important considerations

It is important to note that this study is not a forecasting exercise; the models were designed specifically to analyze market mechanisms rather than to predict future prices. The researchers intentionally excluded lagged price data to focus on fundamental drivers like the power generation mix and cross-border flows. Additionally, while the models capture complex nonlinear relationships better than traditional linear regression, the results are limited by the specific three-year dataset used, which was chosen to avoid the market irregularities caused by the COVID-19 pandemic and the energy crisis following the invasion of Ukraine.

Comments (0)

No comments yet

Be the first to share your thoughts!