Abstract: To address the issues of insufficient control accuracy and the limited manual optimization capabilities of deoxidation alloying during LF refining furnace refining process, an intelligent deoxidation alloying model is developed based on the metallurgical reaction mechanism and historical production data. This model integrates theoretical analysis and data-driven methods, comprehensively considering the alloy elements yield, market prices, and dynamic changes during the production process. It uses optimization algorithms to calculate the amount of alloy addition that yields the highest cost-effectiveness, thereby achieving precise control of deoxidation alloying and cost optimization. In the model design, the metallurgical mechanism part takes into account the dissolution and oxidation processes of alloy elements in the molten steel, providing a theoretical basis for the prediction of alloy elements yield. Meanwhile, the data-driven historical part uses linear regression analysis to extract the quantitative relationship between key operating parameters and alloy elements yield, compensating for the limitations of mechanism model in complex working conditions. Based on this, the model could dynamically adjust the amount of alloy addition to adapt to the fluctuations in the production process and ensure the stability of alloying effect. The results show that the prediction accuracy of this model for ferro-manganese and ferrochrome within±20 kg deviations reaches 90% and 96% respectively, and the accuracy of ferrosilicon deoxidation calculation within±20 kg deviations also exceeds 90%. The successful application of this model not only improves the efficiency and accuracy of deoxidation alloying in LF refining furnace, but also provides technical support for intelligent production in steel enterprises. Through precise control of the alloying process, enterprises could achieve efficient utilization of resources and significant cost optimization, providing important references for the green and intelligent transformation of steel industry.