新型冲击响应谱发生器试验参数智能协同优化设计

Design and intelligent collaboration optimization for testing parameters of a new shock response spectrum generator

为扩充垂直冲击试验机的功能,设计了一种新型冲击响应谱发生器。结合Mindlin板理论与有限元仿真,分析了简支矩形厚板固有特性、激励载荷同冲击谱的关系。提出正交试验、支持向量机与多种群遗传算法结合的智能协同优化方案。利用Nastran软件建立仿真冲击谱的正交试验数据库,通过支持向量机建立仿真冲击谱与目标冲击谱的均方根误差预测模型,采用多种群遗传算法优化谐振板尺寸(板长a,板宽b,板厚h)与激励载荷(峰值A,脉宽D),最终进行仿真与试验验证。优化后的谐振板尺寸为a=398m,b=387mm,h=32mm,激励载荷为A=4.628kN,D0=0.8ms,预测均方根误差为27.7。优化结果表明仿真冲击谱与试验冲击谱均满足冲击试验规范条件,仿真均方根误差为39.6,试验均方根误差为33.4。该研究为冲击试验提供了一种新思路。

In order to expand the function of a vertical impact testing machine,a new type of shock response spectrum(SRS)generator was designed in this paper.Combined with the Mindlin plate theory and the finite element simulation,the relationship among the inherent characteristics of a simply supported plate,the excitation load,and SRS was analyzed.An intelligent collaboration optimization scheme was proposed,which contains the orthogonal test,support vector machine(SVM),and the multi-population genetic algorithm(MPGA).First,we established a database of simulated SRSs by emulating orthogonal tests with Nastran software.Then,we used SVM to establish the surrogate model for predicting the root mean square error(RMSE)of simulated SRSs and the standard ones.Lastly,the optimal parameters of the resonant plate size(length a,width b,thickness h)and excitation load(peak A,width D)were obtained from the surrogate model by MPGA.The optimized results are that the size of the resonant plate is a=398 m,b=387 mm,h=32 mm and the excitation load is A=4.628 kN,D 0=0.8 ms,and the RMSE is 27.7.Meanwhile these optimized results were verified and validated with simulation and experiments.The results show that both the simulated SRS and the experimental SRS meet the requirement of test standard.The simulated RMSE is 39.6 and the experimental RMSE is 33.4.This study provides a new way for impact test.