基于自适应混合近似模型的顶置武器站多柔体系统动力学优化研究

A study on flexible multi-body system dynamics optimization of an overhead weapon station based on an adaptive hybrid approximation model

针对顶置武器站结构优化设计存在计算量大、优化效率低等问题,提出一种基于自适应混合近似模型的优化策略,引入分层设计空间缩减思想,在优化迭代过程中依次在构造的全局空间、聚类空间和重点空间内选取样本点更新混合近似模型,以同时提高模型的全局和局部预测能力。使用典型测试函数算例和某顶置武器站结构动力优化实例,验证了所提优化策略的有效性。顶置武器站结构动力优化结果表明:使用该方法获得的武器站炮口扰动目标函数减小了58.3%,各炮口扰动参数得到有效改善;与静态近似模型方法相比,该方法所得的炮口扰动目标函数优化结果降低了14.5%,所需调用武器站分析计算模型次数减少了47.4%。

An optimization strategy using an adaptive hybrid approximation model was proposed for the problems of large computational complexity and low optimization efficiency in the structural optimization of an overhead weapon station.The idea of layered design space reduction was introduced,and in the optimization iteration process,the hybrid approximation model was updated by successively selecting sample points in the constructed global space,cluster space and key space to improve the global and local prediction capabilities of the model.The effectiveness of the proposed optimization strategy was verified by using a typical test function example and a structural dynamic optimization example of the overhead weapon station.Results show that,the objective function of the muzzle disturbance of the weapon station obtained by this design is decreased by 58.3%,and the muzzle disturbance parameters are effectively improved.Furthermore,compared to the static approximation model algorithm,the objective function of the muzzle disturbance is reduced by 14.5%,and the times of calling the weapon station analysis calculation model are reduced by 47.4%.