MVLFLM: A parameter-efficient large language model framework for cross-domain multi-voltage load forecasting in smart grids

Modern smart grids face significant challenges in short-term load forecasting due to increasing complexity across transmission, distribution, and consumer levels. While recent studies have explored large language models for load forecasting, existing methods are limited by computational overhead, voltage-level specificity, and inadequate cross-domain generalization. This paper introduces Multi-Voltage Load Forecasting Large Model (MVLFLM), a unified Transformer-based framework that addresses multi-voltage STLF through parameter-efficient fine-tuning of a Llama 2-7B foundation model. Unlike previous LLM-based forecasting methods that focus on single voltage levels or require extensive retraining, MVLFLM employs selective layer freezing to preserve pre-trained knowledge while adapting only essential parameters for load pattern recognition. Comprehensive evaluation across four real-world datasets spanning high (transmission), medium (distribution), and low (consumer) voltage levels demonstrates MVLFLM’s superior performance, achieving higher performance than benchmarks. Most significantly, MVLFLM exhibits exceptional zero-shot generalization with only 9.07% average performance degradation when applied to unseen grid entities, substantially outperforming existing methods. These results establish MVLFLM as the unified, computationally efficient solution for multi-voltage load forecasting that maintains forecasting accuracy while enabling seamless deployment across heterogeneous smart grid infrastructures. • Unified Multi-Voltage Framework: the unified comprehensive LLM-based framework for simultaneous STLF across high, medium, and low voltage levels, addressing the scalability limitations of existing single-domain approaches. • Parameter-Efficient Architecture: introduction of a novel PEFT strategy utilizing selective layer freezing that reduces training parameters by 99.2% while maintaining superior forecasting accuracy, enabling practical deployment in resource-constrained environments. • Enhanced Zero-Shot Generalization: demonstration of robust cross-domain forecasting capabilities with minimal performance degradation when applied to previously unseen grid entities, significantly outperforming existing methods. • Comprehensive Multi-Dataset Validation: rigorous evaluation across four diverse real-world datasets representing different voltage hierarchies, temporal resolutions, and geographic regions, providing the most comprehensive assessment.