摘要
大型起竖系统是复杂的非线性系统,难以用统一的数学模型表示,因此用Simulink中的液压仿真工具箱SimHydraulics和机械仿真工具箱SimMechanics搭建了起竖系统实物仿真模型,通过仿真获得了辨识数据,然后用改进的BP神经网络进行了模型辨识。验证结果表明,该模型能够预测起竖系统的输出。综合运用实物仿真和神经网络辨识的方法,为大型起竖系统的优化设计和控制策略的验证提供了参考。
The large erecting system is a complicated nonlinear system, and it is difficult to build its uniform mathematics model. So the physical simulation model of the erecting system was built by hydraulics simulation toolbox SimHydraulics and mechanism simulation toolbox SimMechaics based on Simulink. By simulation data, the modelwas identified by the improved BP neural networks. The results indicate that the BP neural networks model can be used to estimate the output of the erecting system. The method in which physical simulation integrates, with neural networks identification provides reference for the optimal design and control method validation of the erecting system.
出处
《机床与液压》
北大核心
2013年第16期147-150,共4页
Machine Tool & Hydraulics
关键词
起竖系统
非线性系统
BP神经网络
模型辨识
Erecting system
Nonlinear system
BP neural networks
Model identification
