Parameter Calibration for Detonation Model of Explosives Introducing Artificial Intelligence Algorithm

ABSTRACT

This paper proposes a novel parameter calibration method for the detonation model of explosives, incorporating a hybrid particle swarm optimization algorithm (HPSO). We calibrated the three components that together supplement the detonation model: the Jones–Wilkins–Lee (JWL) equation of state for unreacted explosive, the reaction rate equation, and the JWL equation of state for detonation products. When calibrating the parameters of the JWL equation of state for unreacted explosive and the reaction rate equation, the root mean square error between the simulated values and the experimental values of the take‐off speed, take‐off time, and maximum speed of particles at multiple test points inside the explosive was defined as the objective function. When calibrating the parameters of the JWL equation of state for detonation products, the root mean square error between the simulated values and the experimental values of the take‐off speed, take‐off time, and maximum speed of the copper plate was defined as the objective function. If the objective function value was less than 15%, the optimal parameter values were output. The results show that, compared with the manual parameter calibration, the HPSO algorithm can speed up the parameter calibration for these three components by 59%, 7%, and 56%, respectively.