环型聚能装药侵彻靶板能力多因素分析  

Multi-Factors Analysis of Annular Shaped Charge Capability of Penetrating Target Board

作　　者：马焱[1] 赵捍东[1] 黄鑫[1] 

机构地区：[1]中北大学机电工程学院,山西太原030051

出　　处：《高压物理学报》2017年第4期462-468,共7页Chinese Journal of High Pressure Physics

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。The relationship between the structural parameters and penetration capacity of annular shaped charge is hard to express by analytical mathematical formulas, so it is meaningful to establish a model to describe this relationship using the grey relational analysis theory. In our study,the orthogonal test data were at first processed by the grey correlation theory, transforming the multi-objective problem into a single objective problem and obtaining the grey correlation degree between the structural parameters and penetration ability. The support vector machine （SVM） network regression models based on the parameter optimization algorithm methods of support vector regression,particle swarm optimazition,and genetic algorithm were then used to forecast the grey correlation degree and predict annular shaped charge capability. The results show that the fitted curve of the SVM regression model based on the genetic algorithm is the most accurate, and can suc- cessfully describe the relationship between the structural parameters and penetrating ability of the annular shaped charge in the orthogonal test. Finally, using a set of data out of the orthogonal test, the SVM prediction model is verified by comparing the predicted results with the results obtained by LS-DYNA software.

关 键 词：装药技术 环型聚能装药 灰色关联理论 支持向量机 参数寻优 

分 类 号：O315[理学—一般力学与力学基础] TJ410.3[理学—力学]