Active Control of Pantograph Based on Prior-information of Catenary

Aiming at the external disturbance and model uncertainty of high-speed railway pantograph system, a neural network-sliding mode active control strategy of pantograph based on catenary prior information was proposed. Taking advantage of the good self-adaptation of sliding mode control to the perturbation and interference of system and the characteristic of neural network approximation to the uncertain nonlinear function, a neural network-sliding mode pantograph active controller was designed. The ideal vertical displacement of the pantograph head is determined according to the periodic distribution of the contact wire and the speed of the train. The uncertain part of the model is approximated by training and learning the pantograph head displacement error and the error change rate to reduce the influence of the uncertain part to the pantograph head displacement. The stability and convergence of the system are guaranteed by adaptive weight adjustment. The control effect of controller on model uncertainty and stiffness change is analyzed through simulation, which validates that the pantograph neural network sliding mode variable structure active control based on catenary prior information proposed in this paper has strong anti-interference to parameter perturbation and make pantograph head displacement and contact wire displacement have better matching characteristic.