基于概率-椭球混合模型的缓冲气囊防护特性可靠性分析方法

Hybrid Reliability Approach for Airbag Protection Performance Based on Probability and Ellipsoidal Convex Models

针对多重不确定性因素对缓冲气囊防护特性的影响,提出一种基于概率-椭球混合模型的缓冲气囊防护特性可靠性分析方法。首先,利用概率模型和椭球凸集模型对装备空降过程中的不确定性因素进行描述,并建立缓冲气囊防护特性可靠性分析模型;然后,利用Karush-Kuhn-Tucker最优化条件将可靠性指标求解时面临的双层嵌套优化问题转化成单层优化问题,实现循环嵌套优化问题的解耦,从而提高可靠性指标的求解效率;最后,利用隔代映射遗传算法对气囊防护可靠性指标进行求解,并通过开展装备落地碰撞试验对缓冲气囊的防护特性进行验证。结果表明,该方法能有效对缓冲气囊的防护特性进行可靠性评估,并确保空降装备的安全性,在装备空降技术领域具有一定的实际工程意义。

Considering the influence of multi-source uncertainties on airbag protection performance,a hybrid reliability approach based on probability and ellipsoidal convex models is presented for airbag protection performance.Firstly,the probability and ellipsoidal convex models are used to quantify the uncertainty,and the reliability analysis model of airbag protection performance will be constructed.Then,considering the original hybrid reliability analysis belongs to a complex nesting optimization problem,an efficient decoupling strategy is suggested to improve the computational efficiency.Based on this decoupling strategy,the normalized hybrid reliability model can be converted into the singer-layer equivalent model by utilizing the Karush-Kuhn-Tucker conditions.Finally,the intergeneration projection genetic algorithm(IP-GA)is adopted to obtained the reliability index of airbag protection performance,which is validated by the vehicle airdrop experiments.The results demonstrate that the proposed method could effectively evaluate the reliability of airbag protection performance and ensure the vehicle safety.The method can also be used in the field of other airdrop protections.