论文部分内容阅读
针对防滑控制系统结构复杂,尤其是影响轮轨间粘着系数的随机因素太多,难以用传统的控制理论建立控制模型,笔者摒弃了常用的速度差参数,而选用滑移率、加减速度和冲动3个物理量作为防滑判据,根据这3个物理量建立了防滑器模糊神经网络控制模型,并开发了相应的仿真软件。笔者通过现车的部分速度试验数据进行了控制模型的验证,结果表明该控制模型能够依据滑移率、加减速度和冲动正确判断车轮的运行状况,控制排风阀做出正确反应,从而达到适时调节制动缸压力的目的。
In view of the complex structure of the anti-skid control system, especially the stochastic factors affecting the adhesion coefficient between the wheel and rail, it is difficult to establish the control model with the traditional control theory. The author abandons the commonly used speed difference parameters and selects slip ratio, acceleration and deceleration Impulsive three physical quantities as a non-slip criterion, based on these three physical quantities to establish a control model of antiskid fuzzy neural network, and the corresponding simulation software developed. The author verifies the control model through the test data of the current car speed. The results show that the control model can correctly judge the wheel running condition based on the slip rate, acceleration and deceleration and impulse, and control the exhaust valve to make the correct response, so as to achieve Timely adjustment of the purpose of brake cylinder pressure.