A Comparative Study of Machine Learning Algorithms for Electricity Price Forecasting with LIME-Based Interpretability

TL;DR

This paper conducts a systematic comparison of eight machine learning models for electricity price forecasting in the Spanish market using 2015–2018 hourly data, integrating consumption, generation, and weather features. It finds nonlinear models, especially KNN, provide the best predictive performance, while linear baselines lag significantly. To address interpretability, the study applies LIME to reveal that meteorological factors and supply-demand indicators drive price fluctuations through nonlinear relationships. The combined forecasting and interpretability framework offers accurate predictions with enhanced transparency for market participants and policymakers.

Abstract

With the rapid development of electricity markets, price volatility has significantly increased, making accurate forecasting crucial for power system operations and market decisions. Traditional linear models cannot capture the complex nonlinear characteristics of electricity pricing, necessitating advanced machine learning approaches. This study compares eight machine learning models using Spanish electricity market data, integrating consumption, generation, and meteorological variables. The models evaluated include linear regression, ridge regression, decision tree, KNN, random forest, gradient boosting, SVR, and XGBoost. Results show that KNN achieves the best performance with R^2 of 0.865, MAE of 3.556, and RMSE of 5.240. To enhance interpretability, LIME analysis reveals that meteorological factors and supply-demand indicators significantly influence price fluctuations through nonlinear relationships. This work demonstrates the effectiveness of machine learning models in electricity price forecasting while improving decision transparency through interpretability analysis.

Figures (5)

• Figure 1: Data Preprocessing Flow for Electricity Price Prediction
• Figure 2: Scatter plot of actual vs. predicted electricity prices.
• Figure 3: Time series comparison of actual and predicted prices (first 100 steps).
• Figure 4: Distribution of prediction errors.
• Figure 5: Feature contributions visualized by LIME.