Research on Real-time Carbon Emission Factor Prediction Method for Load Nodes of Power System based on Gated Recurrent Cell Network

With the development of the economy, the daily use of electricity is also expanding in scale. The form of power generation has also transformed from the traditional forms of hydro and thermal power to wind power and photovoltaic participation in the generation mode. At the same time, load and generation fluctuate more significantly with the seasons, and there are peaks and valleys in the daily load curve. The traditional tide analysis-based carbon flow calculation model requires the collection of information from individual nodes for centralized tide calculation. However, with the complexity of the power system structure and the sharp fluctuations in load and generation, there are situations where information collection is unreliable and the complexity of the calculation is high, and historical data cannot be used to help with carbon emission calculation and analysis. In order to reduce the computational cost, improve the computational stability and enhance the generalization ability of the model, this paper proposes a real-time carbon emission factor prediction method for load nodes of power systems based on gated recurrent cell networks. The method is able to quickly predict the real-time carbon emissions of nodes by analyzing the historical data of node loads and carbon emission factors. Finally, simulation experiments are carried out in the IEEE 9 nodes system and the accuracy of the assessment method is verified.