Research on Multi Energy Coupling Load Forecasting Method for Low Carbon Parks Based on Speed Optimized GRU Model

In response to the national call for dual carbon and the construction of low-carbon parks, there are mainly problems such as a single type of load forecasting, missing typical energy load data, and high difficulty in simulation. At present, research on the multi energy coupling load prediction of low-carbon parks is basically in a blank stage. Continuing the existing system model-based forecasting approach for park load forecasting, introducing machine learning models with higher prediction accuracy and improving optimization speed, a low-carbon park multi energy coupled load forecasting model is proposed. Based on the actual data of the cooling season and heating season of an integrated energy system, the accuracy of the model is verified. The verification shows that the prediction model proposed in this paper is fully progressiveness and the speed of optimization is greatly improved.