<p>The blasting excavation of ore body or rock mass still has great failure risk and high energy consumption problems. In this paper, the blasting characteristic index system of magnetite ore is established and a novel nonlinear blasting energy consumption prediction model based on multiple regression theory is proposed, which provides a basis for reasonable estimation of blasting energy consumption. Based on the split Hopkinson pressure bar, the strength characteristics, granularity parameters, and the evolution law of energy consumption of magnetite ore are studied. The results show that the increase of elastic energy storage limit is the main factor to improve the dynamic compressive strength. The increase of fracture energy causes the crack to spread in a nonlinear way, increases the cross-section and irregularity of the specimen, increases the risk of instability failure of the ore, and leads to the increase of fractal dimension and the decrease of characteristic particle size. The energy response of mineral particle defects and microcracks and the high energy demand of interparticle dissociation are the key factors to increase the proportion of energy absorbed by ore. It is suggested that the incident energy range is 50 ~ 200&#xa0;J, and the energy utilization rate is high in this range, and the blasting effect is better. For blasting excavation, according to the target particle size, strength and component content, the prediction model is used to calculate the blasting energy consumption range of the ore body, and the appropriate adjustment is made according to the quality grade of the rock mass.</p>