Stacked-GRU网络在动力随动陀螺零部件选配上的应用

Selective Assembly for Dynamical Servo Gyro Parts with Stacked-GRU Network

动力随动陀螺是红外导引头位标器的核心部件之一,其装配质量对于导弹制导精度而言有着决定性的作用。针对动力随动陀螺零部件一次选配成功率低的问题,提出了一种基于数据驱动的动力随动陀螺零部件选配方法。对动力随动陀螺零部件装配现状进行了分析,考虑到零部件装配参数间复杂的非线性映射关系,建立了堆叠门控循环单元(Stacked-GRU)网络数据驱动模型,对陀螺待装配零部件参数属性值进行了预测。根据预测结果指导了陀螺零部件的选配,并以陀螺典型零部件装配为例进行了验证。实验结果表明,所提方法能够实现动力随动陀螺零部件的选配,且预测精度优于传统神经网络。

Dynamical servo gyro is one of the core components of infrared seeker coordinator,and its assembly quality plays a decisive role in missile guidance accuracy.Aiming at the low success rate of the first-time matching of dynamical servo gyro parts,a data-driven method for the selective assembly of dynamical servo gyro parts was proposed.The assembly process of dynamical servo gyro parts was analyzed firstly.Considering the complex nonlinear mapping relationship between the assembly parameters,a stacked GRU network data-driven model was established to predict the attribute values of the component to be assembled.According to the prediction results,the selective assembly of gyro parts was guided,and the typical assembly of gyro parts was verified as an example.The experimental results show that the proposed method can realize the selective assembly of dynamic servo gyro parts,and the prediction accuracy is better than the traditional neural network.