A Method for Low Carbon Operation Decision Support of Thermal Power Plants Based on Data-driven Causal Inference

The national carbon market is one of the core policy tools to achieve the carbon peaking and carbon neutrality goals. In 2021, the first performance period of the power generation industry in the national carbon market was officially launched, and the thermal power industry is the only emission control industry in the national carbon market. Under the background of the current dual carbon revolution, emission control requirements are gradually increased, and the impact on the economic efficiency of thermal power plants is further amplified by whether low-carbon operation can be realized. However, due to the impact of technology and other objective factors, the carbon emission intensity of thermal power units is affected by the unit characteristics, fuel, load rate and other factors, and there are many factors affecting the carbon emission level of thermal power plants. Therefore, how to achieve reasonable control of thermal power plant carbon emissions under given conditions becomes an important factor affecting the economics of thermal power plants. Since various operational control quantities of thermal power plants, such as power supply and heat supply, are closely related to carbon emissions, it is effective to construct a cause-effect diagram to fit the relationship between operational control quantities and carbon emissions based on data-driven cause-effect inference, and to perform sensitivity analysis on carbon emission levels to achieve reasonable control of carbon emissions for low-carbon operation of thermal power plants. In this paper, we present a data-driven causal inference method for low-carbon operation of thermal power plants. The causal diagram is constructed through a priori knowledge, and the relationship equation between the intervening variables and the outcome variables is fitted based on the matching model, and the sensitivity analysis of the unit carbon emissions is carried out to realize the auxiliary decision-making for low-carbon operation of thermal power plants.