Research on Short Term Prediction Technology of Regional Air Conditioning Based on Time Series Large Model

In order to solve the problem of insufficient data volume and low prediction accuracy caused by small sample data This article proposes a time series prediction large model based on Prompt+large language model. Firstly, Spearman is used for correlation analysis to obtain strong correlation factors that affect air conditioning changes. Secondly, the data is smoothed using a cubic weight function to reduce data noise. On the basis of this, design the UniTS Prompt time series large model structure again, and use the grey wolf optimization algorithm to optimize the hyperparameters of the network Finally, the limitations of the proposed method were verified through numerical examples.