Mudflow disasters can occur easily in loess regions, due to the particular characteristics of loessic material. Such disasters have caused considerable damage to both the physical and the human environment. Little research has been conducted into the dynamic pressure caused by the different densities and weights of different debris flows in loess regions. A flume test was therefore built to simulate the impact pressure of debris flows, and the peak impact pressure was estimated. Moreover, hydrodynamic and hydrostatic models were improved using these experimental results. Finally, by combining them with the dimensionless analysis and Buckingham π theorems, the Froude number and the Reynolds number were introduced to construct a comprehensive dimensionless relationship equation for debris flows.
The results showed that the debris flow impact process was divided into three phases: the first phase was characterized by a rapid and powerful impact on the sensors; the second stage saw a continuous and steady impact on the sensors’ supports; and the third and final phase was very stable. In this study, the velocity was 1.23 ~ 3.62 m/s when the density of the debris flow ranged from 1,100 kg/m3 to 2,300 kg/m3 and the weight of mixture was 100 kg ~ 500 kg, that the flow depth was 2.7 ~ 13.4 cm, and that the peak impact pressure ranged between 1.23 and 28.41 kPa. These results would suggest that the density and weight of the debris flow exerts a significant impact upon any experimental results.
In addition, the hydrodynamic and hydrostatic models were modified; the adjusted values these models yielded were > 0.721. Specifically, the modified hydrodynamic model was observed to perform very well with high velocities and comparatively high Froude numbers, but the modified hydrostatic model was relatively appropriate for low velocities and low Froude numbers. The peak dimensionless pressure was the expressed as a power function of the Froude number and the Reynolds number, respectively. The comprehensive peak dimensionless pressure for debris flows was subsequently coupled and expressed as a power function, and the adjusted value (R2) was 0.803.
These results would indicate that the modification model and the comprehensive can both be applied to the loess regions of China, and can provide a better understanding of solid–liquid two phase flow mechanisms, as well as support engineering design work and risk assessments in loess regions affected by debris flows.