The bottom blown oxygen copper smelting technology is a new copper concentrate smelting technology. However, in the actual production, it is difficult to measure some key parameters online such as the matte grade, slag Fe/SiO2, melt temperature, total level, and matte level. A dynamic model is established to predict these parameters online, and is verified using actual production data. The results show that the predicted data and measured data are close, which indicate that the model could effectively predict these key parameters online and be applied to the actual production.