Knowledge Mapping from the Dynamic Coupling Mechanism of Load Forecasting Error-operation Cost-Carbon Emissions Based on Multi-dimensional Collaborative Optimisation

Under the global goal of energy conservation and emission reduction, the power system needs to balance economy and low carbon emissions. The error of load prediction directly affects the formulation of power generation plan. Most of the existing studies analyze load forecasting error, operation cost and carbon emissions(ECE) in isolation, and lack a quantitative model for the dynamic coupling relationship between the three. This theoretical gap makes it difficult for traditional dispatching strategies to balance the contradiction between economy and low carbon emissions in the new power system. This study develops a dynamic coupling optimisation framework of ‘ECE’, which embeds an optimal error prediction model in a trinomial dynamic coupling equation to construct a dynamic feedback mechanism. In the first stage of building the dynamic coupling optimisation framework of ‘ECE’, the optimal error prediction model is found to be ‘random forest’ by analysing four types of typical national data. In the second stage, the predicted error value is substituted into the three-base dynamic coupling equation. The dynamic coupling model can output important results such as the prediction of the effect of policy interventions and the differences in coupling patterns between countries. This study establishes, for the first time, a dynamic coupling equation of ‘ECE’ under the guidance of policy indicators, revealing the threshold law of error transmission.