对影响驼峰机车推送速度的各种因素进行了探讨,将模糊神经网络理论引入到驼峰变速推峰系统中,结合神经网络和模糊控制的各自优点,将专家或熟练操作人员的控制策略转换为控制功能函数,利用神经网络自学习、自校正的能力调整、修改模糊控制规则,从而实现对推送速度的优化控制。 Various factors affecting the speed of hump locomotive pushing are discussed. The theory of fuzzy neural network is introduced into hump peak pushing system. Combining the respective advantages of neural network and fuzzy control, the control strategy of expert or skilled operator is converted into control Function function, the use of neural network self-learning, self-tuning ability to adjust, modify the fuzzy control rules, in order to achieve the optimal control of the push speed.