A Few-Shot LLM Framework for Extreme Day Classification in Electricity Markets

2026-02-17

AI Summary

This study addresses the challenge of forecasting extreme day-ahead electricity price spikes in power markets, particularly under data-scarce conditions. The authors propose a novel few-shot classification approach leveraging large language models (LLMs). System states—including electricity demand, renewable generation, weather forecasts, and recent price trends—are transformed into statistical features and encoded as natural language prompts fed into the LLM, which then classifies peak-price days and provides confidence scores. To the best of the authors’ knowledge, this is the first application of few-shot LLMs to electricity price spike prediction. Evaluated on Texas market data, the method matches the performance of SVM and XGBoost when ample labeled data are available and significantly outperforms them under limited labeling, thereby reducing reliance on extensive annotated datasets and demonstrating the potential of LLMs for data-efficient power market analysis.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

This paper proposes a few-shot classification framework based on Large Language Models (LLMs) to predict whether the next day will have spikes in real-time electricity prices. The approach aggregates system state information, including electricity demand, renewable generation, weather forecasts, and recent electricity prices, into a set of statistical features that are formatted as natural-language prompts and fed to an LLM along with general instructions. The model then determines the likelihood that the next day would be a spike day and reports a confidence score. Using historical data from the Texas electricity market, we demonstrate that this few-shot approach achieves performance comparable to supervised machine learning models, such as Support Vector Machines and XGBoost, and outperforms the latter two when limited historical data are available. These findings highlight the potential of LLMs as a data-efficient tool for classifying electricity price spikes in settings with scarce data.

Problem

• electricity price spikes
• few-shot learning
• extreme day classification
• data scarcity
• electricity markets

Innovation

• Few-shot learning
• Large Language Models
• Electricity price spike classification
• Data-efficient prediction
• Natural-language prompting