Contact Load Control in Pantograph-Catenary Based on Wavelet Neural Network Backstepping Method

With the rapid development of electrified railroad, the pantograph-catenary system’ s(PCS) current-carrying performance has become a crucial factor limiting the safety and dependability of high-speed train operation. From the perspective of control theory, an adaptive backstepping controller based on Wavelet Neural Network(VVNN) is proposed to stabilize the contact force and improve the current-carrying performance of the system. Specifically, the virtual control laws are adopted, Lyapunov functions are selected by stepwise inversion recursion, the unknown parts of the model are approximated by WNN separately, and the errors generated by approximation of WNNs are compensated by introducing robust term. The parameter adaptive law for the WNN and the update rule for the robust term are obtained by derivation of Lyapunov function. The corresponding stability analysis is also carried out. The simulation results show that the backstepping control strategy based on WNN can compensate the influence of uncertainty well, reduce the fluctuation of the contact load of the PCS effectively and improve the current-carrying performance.