环型聚能装药侵彻靶板能力多因素分析

发布时间：2018-03-27 08:21

本文选题：装药技术　切入点：环型聚能装药　出处：《高压物理学报》2017年04期

【摘要】：环型聚能装药结构参数与其侵彻靶板能力间的关系难以用精确的数学函数表达,因此利用灰色关联理论建立描述该关系的模型是有意义的。首先采用灰色关联度理论对正交试验数据进行初步处理分析,将多目标问题转化为单目标问题,得到各结构参数与侵彻靶板能力的灰色关联度;然后应用基于支持向量机回归、粒子群优化、遗传算法等参数寻优算法的支持向量机(SVM)网络回归模型对灰色关联度进行预测,从而实现对环型聚能装药侵彻靶板能力的计算。结果表明,使用基于遗传算法参数寻优的SVM网络回归模型拟合精度最高,该模型可以很好地描述正交试验中环型聚能装药结构参数与侵彻靶板能力间的关系。最后选用正交试验外的一组数据,应用LS-DYNA对该结构参数下的环型聚能装药侵彻靶板过程进行仿真,并将仿真试验数据与SVM网络回归模型的预测值作比较,验证了该模型的可靠性。
[Abstract]:The relationship between ring shaped charge structural parameters and penetrating target ability to use mathematical function accurately, so using grey correlation theory to describe the relationship between the model is meaningful. Firstly, by using the grey correlation theory of analysis on the data of orthogonal test, the multi-objective problem is transformed into a single objective problem. Grey correlation degree obtained the structural parameters and penetrating target ability; then using support vector machine regression based on particle swarm optimization, SVM parameter optimization algorithm and genetic algorithm (SVM) to predict the grey relational network regression model, so as to realize the calculation of the charge penetration target of ring shaped the use of SVM network. The results show that the genetic algorithm parameters optimization of regression models based on the highest accuracy, the model can well describe the orthogonal test in shaped charge structure. The relationship between the number and the penetrating target ability. The final selection of a set of data of orthogonal test of simulation parameters on the structure and application of ring shaped LS-DYNA charge penetrating the target process, and simulation test data and the SVM network models compared, verified the reliability of the model.

【作者单位】：中北大学机电工程学院;
【分类号】：O385;TP18

【相似文献】