Real-time accurate prediction of multi-ferrite quality in blast furnace ironmaking is an important means for real-time monitoring and effective control of blast furnace internal state, but the prediction results are not accurate and lack credibility, especially in unstable furnace conditions and in the case of large fluctuations in blast furnace data, the prediction results of multi-component molten iron quality have large deviations and low credibility, and cannot provide guidance for daily operation and adjustment of blast furnace. Aiming at this engineering problem, this paper proposes a Stacking-based blast furnace hot metal quality interval prediction modeling algorithm for establishing the molten iron quality interval prediction model. Firstly, in order to improve the accuracy of modeling, a Stacking-based hot metal quality model is established. Then, in order to characterize the credibility of the prediction results, an interval prediction method is introduced to simultaneously predict the multi-ferrite water quality value and the prediction interval. Finally, industrial experiments show that the proposed algorithm can accurately predict the quality of blast furnace multi-metal hot metal, and the prediction interval has high credibility, which has a good guiding significance for actual production operations.