复合装药安全性优选BP神经网络模型

为在设计复合装药结构时得到复合装药的能量输出和安全性双优结构,利用BP人工神经网络理论,建立结构优选模型。利用不同的感度和能量输出试验数据,进行模型的训练,训练后对样本进行预测和优选。结果表明,该方法能够克服评估复合装药结构性能优劣的各因素之间的复杂关系,实现精确的非线性预测,得到安全性和能量输出双优的设计结构。

An Optimized BP Neural Network Model for the Selection of Composite Charge Structures

For the purpose of obtaining the charge structure with low sensitivity and high energy output during the design of composite charge structures,an optimized selection model was proposed based on the BP neural network theory.The model was trained by using the test data with different sensitivities and energy outputs.And then a selection was made on the trained samples and the best optimized model can be selected out.The result shows that the proposed method can overcome the complex relationships between diffe...