M2WLLM: Multi-Modal Multi-Task Ultra-Short-term Wind Power Prediction Algorithm Based on Large Language Model

2025-05-31

AI Summary

To address the low accuracy and poor generalizability of ultra-short-term wind power forecasting, this paper pioneers the integration of large language models (LLMs) into wind power prediction. We propose a multimodal, multi-task joint modeling framework. Key innovations include a Prompt Embedder and a semantics-enhanced Data Embedder, enabling semantic alignment and joint encoding of textual prompts and heterogeneous time-series data; and a Semantic Augmenter to improve few-shot adaptability and cross-regional robustness. Evaluated on real-world wind power datasets from three provinces, our method achieves an average 18.7% reduction in MAE over state-of-the-art baselines—including GPT4TS—across 5–30-minute prediction horizons. The results demonstrate substantial improvements in both forecasting accuracy and stability, offering reliable support for real-time power grid dispatch and resource optimization.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• 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

The integration of wind energy into power grids necessitates accurate ultra-short-term wind power forecasting to ensure grid stability and optimize resource allocation. This study introduces M2WLLM, an innovative model that leverages the capabilities of Large Language Models (LLMs) for predicting wind power output at granular time intervals. M2WLLM overcomes the limitations of traditional and deep learning methods by seamlessly integrating textual information and temporal numerical data, significantly improving wind power forecasting accuracy through multi-modal data. Its architecture features a Prompt Embedder and a Data Embedder, enabling an effective fusion of textual prompts and numerical inputs within the LLMs framework. The Semantic Augmenter within the Data Embedder translates temporal data into a format that the LLMs can comprehend, enabling it to extract latent features and improve prediction accuracy. The empirical evaluations conducted on wind farm data from three Chinese provinces demonstrate that M2WLLM consistently outperforms existing methods, such as GPT4TS, across various datasets and prediction horizons. The results highlight LLMs' ability to enhance accuracy and robustness in ultra-short-term forecasting and showcase their strong few-shot learning capabilities.

Problem

• Accurate ultra-short-term wind power forecasting for grid stability
• Overcoming traditional methods' limitations with multi-modal LLM integration
• Improving prediction accuracy via textual and numerical data fusion

Innovation

• Leverages Large Language Models for wind prediction
• Integrates textual and numerical data multi-modally
• Uses Prompt and Data Embedders for feature fusion