A data assimilation method for blast load prediction

Abstract

Blast load is a crucial parameter in blasting construction design and numerical simulation. However, it is characterized by dense energy, high loading rate and complex action process, which makes its accurate measurement difficult. In this study, based on experimental data-driven physical simulation, a data assimilation method for blast load prediction was proposed by integrating PFC2D software and Python language. The method can accurately measure the blast load, as it achieves an exact match between the numerical calculation results and the test far-field strain waves by automatically adjusting the peak blast pressure, rise/fall time and rock damping ratio. Furthermore, the effectiveness of the method was verified by performing rock blasting disturbance tests, and the superiority of the data assimilation algorithm was illustrated from two aspects, i.e., time-loading path and rock damping ratio. The following research results were obtained: 1) When the blast pressure is constant, the number of micro cracks within the rock decreases first and then increases with the rise of loading rate, while it always increases with the rise of unloading rate. Therefore, it is necessary to measure the time-loading path of blast pressure. 2) When the rock damping ratio deviates more notably from the true value, the error rate of the peak blast pressure by the assimilation method is greater correspondingly, indicating that the rock damping ratio is a prerequisite for accuracy of the assimilation method. 3) Superior to the traditional isentropic expansion theory, the assimilation algorithm not only accurately measures the peak blast pressure but also considers both the time-loading path and the rock damping ratio. The research results enrich rock dynamic tests and provide valuable references about blasting engineering.