基于智能模糊控制在机车黏着优化策略中的研究

Research on the Optimization Strategy of Locomotive Adhesion Based on Intelligent Fuzzy Control

在机车运行过程中,轮轨间的黏着条件受天气、车辆结构等影响复杂多变。黏着性能好坏直接决定机车牵引力能否正常发挥。传统的黏着优化一般采用组合矫正等方法,该类方法结构简单,存在自适应能力弱、黏着利用率偏低等问题。文中提出一种新型的黏着优化控制方法,它是将专家知识和试验数据采集的样本通过训练学习,建立一种具有自适应能力的T-S模糊神经网络控制器,来寻找最优黏着控制点提高黏着的利用率。该方法无需建立精确数学模型、能够自适应各种复杂多变的黏着条件,具有很强的鲁棒性。最后通过仿真试验验证,文中采用智能模糊控制技术在机车黏着优化策略中,能够明显提高机车牵引力的发挥,改善车辆的控制性能。

In the process of locomotive operation,the adhesion condition between wheel and track is affected by weather and vehicle structure.Adhesion performance directly determines whether the locomotive traction can play normally.The traditional adhesion control adopts the correction method,which has the advantages of simple structure,weak adaptive ability and low adhesion utilization.In this paper,a new adhesive optimal control method is proposed.It is to establish a T-S fuzzy neural network controller with adaptive ability by training and learning the samples collected from expert knowledge and test data,so as to find the optimal adhesive control point and improve the adhesive utilization rate.This method does not need to establish accurate mathematical model,can adapt to various complex and changeable adhesion conditions,and has strong robustness.Finally,through the simulation test,this paper uses the intelligent fuzzy control technology in the locomotive adhesion optimization strategy,which can significantly improve the performance of locomotive traction and improve the control performance of the vehicle.