论文部分内容阅读
采用炸药的氧平衡为影响因素,以炸药的芳香性、α-CH键等因素为控制符建立炸药撞击感度的人工神经网络预测模型。采用遗传算法进行人工神经网络模型的权值、阈值优化,得到更优化的人工神经网络模型,实现了炸药爆炸性能的预测。预测结果表明:预测值与文献值误差较小,说明可以采用人工神经网络模型进行炸药撞击感度的预测,采用遗传算法优化后的遗传-神经网络模型预测误差更小,效果更佳。
Adopting the oxygen balance of explosives as the influencing factor, an artificial neural network prediction model was established to establish the impact sensitivity of the explosives based on the aromaticity of the explosives, α-CH bond and other factors. The weight and threshold of the artificial neural network model are optimized by using genetic algorithm, and a more optimized artificial neural network model is obtained to predict the explosive performance of the explosive. The prediction results show that the error between the predicted value and the literature value is small, indicating that artificial neural network model can be used to predict the impact sensitivity of explosives. The GA-optimized genetic-neural network model has smaller prediction error and better effect.