Power load prediction method based on large language model

The invention belongs to the technical field of power load prediction and the field of natural language processing in a computer, and discloses a power load prediction method based on a large language model, which utilizes the powerful natural language capability and rich pre-training knowledge of the large language model and combines power load related data in a specific scene to predict the power load. According to the method, structured power load data containing multiple influence factors are converted into corresponding text description through a prompt word template, a large language model learns data knowledge in a power load prediction scene on the basis of texts through a supervised efficient fine tuning method, and in the prediction stage, the power load prediction efficiency is greatly improved. And the user informs the large language model of the influence factors by using a natural language, and the model can give a power load prediction result. The method has higher prediction accuracy and does