Given the constraints of the “dual carbon” goals and “dual control” policies

accurately analyzing spatiotemporal differences in electricity carbon emissions and evaluating the low-carbon operation levels of power systems are crucial. This study proposes a new evaluation method for low-carbon operation in power systems

considering spatiotemporal differentiation. By normalizing statistical indicators across spatiotemporal scales

it addresses the limitations in indicator applicability. A multi-dimensional evaluation indicator system is constructed based on the carbon emission factor method and carbon emission flow theory

reflecting the impacts of power structure transitions

grid operations

and policy regulation. The random forest algorithm is used to optimize indicator weights

and the technique for order preference by similarity to ideal solution (TOPSIS) method is applied for comprehensive evaluation. Applications show the method effectively reflects low-carbon operation performance

enables spatiotemporal comparisons

and supports decision-making for low-carbon management in new-type power systems.