基于电枢运行过程中所受到的多物理场作用机理,采用灰色理论对出口速度进行线性预测,分析了灰色理论存在的不足和改进措施,提出了一种Elman动态灰色神经网络并进行了仿真分析,然后将结果与传统的灰色神经网络结果进行了对比试验。结果表明:新的Elman动态神经网络预测方法具有较高的预测精度,可以为轨道式电磁发射系统连发决策、有效预防轨道烧蚀提供技术支撑。
Based on the action of the multi-physical field where the armature of the system works,the muzzle velocity is predicted by the linear prediction method.First of all,the grey model′s disadvantages and its improvement are analyzed.Then a new Elman dynamic grey neural network is presented for high accuracy prediction.Finally,the new model is simulated and experimented together with the traditional grey neural network method.The results show that the new-type prediction method is more accurate than the traditional grey-neural-network prediction method.It provides technical support for the proper operation of the system,the prediction of its working state and the prevention of rail erosion.