Neural Network Supervised Control Based on Levenberg-Marquardt Algorithm
Zhao Hong¹,Zhou Ruixiang²,Lin Tingqi¹
(1.School of Mechanical Engineering,Xi'an Jiaotong University,Xi'an 710049,China;2.The Engineering Institute of the Airforce Engineering University)

Abstract£ºA multilayer neural network supervised on¡Mline control strategy based on Levenbergª²Marquardt training algorithm is proposed for the tracking control problem of the electro-hydraulic position servo systems subjected to constant and time-varying external load disturbances. The Levenberg-Marquardt algorithm is the combination of the steepest decent algorithm with the Gauss-Newton algorithm. Compared with a conjugate gradient algorithm and a variable learning rate algorithm, the Levenberg-Marquardt algorithm is much more efficient than either of them on the training steps and accuracy. Therefore, it can be applied to online control. The output of the system successfully tracked the specified sinusoidal after a relatively short online training period. The control strategy is used to adapt to uncertainties of disturbances and learns their inherent nonlinearities. Simulation results illustrate that a neurocontroller used in supervised control schemes can result in good robustness and tracking property.
Keywords:neural network;electroª²hydraulic position servo systems;supervised control