Application of neural networks in modeling of the transmission hydraulic actuator

摘要：

The basic function of automotive transmission is to transfer the engine torque to the vehicle with the desired ratio smoothly and efficiently.In the automatic transmission systems,the shifting processes are mainly controlled by electro-hydraulic actuators.The availability of pressure information of a hydraulic actuator makes it possible to improve the fuel economy,reduce emission and enhance driving performance.There are many factors that influence the responses of the hydraulic actuator.Most of the models of the actuators are based on Newton’s second law and didn’t consider all of the factors that may influence the feature of the actuators significantly.The simplified process model may cause the inaccuracy of the control performance.So it is very hard to get the accurate model that reflects the nonlinear character of the hydraulic actuator through the routine methods.Neural networks are quite useful in modeling of the routine nonlinear systems; they are very convenience and accuracy.But the returning parameters of the networks are the weights and bias,they have no actual sense but some numeric values.This paper presents a Radial Basis Function (RBF) neural network based algorithm to estimate the parameters of the hydraulic actuator in a vehicle power transmission control system.The trained results of the neural networks are the parameters of the transfer function of the hydraulic actuator that reflect the physical features.By means of experiments,it is shown that the proposed model can describe the main phenomena characterizing of the hydraulic actuator dynamics very well.

关键词： transmission hydraulic actuator model radial basis function (RBF) neural networks transfer function parameters identification.

作者: Tao-tao Jin Ping-kang Li

作者单位: School of Mechanical,Electronic and Control Engineering,Beijing Jiaotong University,Beijing,China

会议类型: 国际会议

会议名称: International Conference on Modelling,Identification and Control(模拟、鉴定、控制国际会议)

会议地点: 上海

会议语种:英文

在线出版日期: 2008-06-29（万方平台首次上网日期，不代表论文的发表时间）