Enhanced Thermal Power Load Data Prediction Based on Large Models

Thermal power load forecasting plays an important role in the system operation and optimized allocation of cogeneration plants, and large models have significant advantages in the field of time series forecasting. The use of large-scale models for thermal and electric load forecasting has important research value and potential. Therefore, for the thermal power load dataset, we have designed a data augmentation method and improved the large model to enhance the capture ability of thermal power load data features. Experimental results have shown that our data augmentation method and improved large model have good predictive performance on the dataset, with a MAPE of 0.9%, fully demonstrating the effectiveness of our method and the development potential of large models in time series prediction problems.