燃料空气炸药爆轰产物JWL状态方程参数计算

炸药爆轰产物的Jones-Wilkins-Lee(JWL)状态方程参数一般由圆筒试验确定，但圆筒试验并不适用于宏观上呈云雾状态的燃料空气炸药(FAE)。为确定FAE爆轰产物的JWL状态方程参数，基于外场FAE爆轰试验数据，引入反向传播神经网络联合遗传算法(BPNN-GA)，建立适用于FAE的状态方程参数计算方法，并与单爆源和多爆源的外场试验结果对比。研究结果表明：引入BPNN-GA可以简化状态方程参数优化过程，提高了寻优速度和精度；基于FAE爆轰产物JWL状态方程参数，建立单爆源与多爆源的FAE云雾爆轰模型，数值仿真所得的冲击波轮廓与实际爆轰冲击波形貌一致，单爆源与多爆源50 m测点处地面峰值超压仿真值与试验值的最大偏差分别为9.0%和11.1%.

Parameters Calculation of JWL EOS of FAE Detonation Products

The JWL EOS parameters of detonation products for high explosives are generally determined by the cylinder test. However, the cylinder test is not suitable for fuel air explosives (FAE), which is cloud-like in a macro state. A method for calculating the EOS parameters based on the experimental data of FAE detonation in external field is established to determine the JWL EOS parameters of FAE detonation products. A back propagation neural-based genetic algorithm (BPNN-GA) is introduced into the method. The calculated values are compared with the data from the single- and multi-source external field experiments. The research shows that the introduction of BPNN-GA can simplify the EOS parameter optimization process and also improve the speed and accuracy. Based on the obtained JWL EOS parameters of FAE, the single- and multi-source FAE cloud detonation models are established. The profile of shockwave front from the simulation is consistent with the morphology of actual detonation shockwave. The maximum deviations between simulated and experimental values of the ground peak overpressure at the 50 m measuring points from single- and multi-source are 9.0% and 11.1%, respectively.