A feedforward neural network fuzzy grey predictor-based controller for force control of an electro-hydraulic actuator

Recently, electro-hydraulic actuators (EHAs) have shown significant advantages over conventionally valve-controlled actuators. EHAs have a wide range of applications where force or position control with high accuracy is exceedingly necessary. Besides the benefits, however, EHA is a highly complex nonlinear system which causes challenges for both modeling and control tasks. This paper aims to develop an effective control approach for force control of a typical EHA system. This control scheme is considered as an advanced combination of a feedforward neural network - based PID (FNNPID) controller and a fuzzy grey predictor (FGP), shortened as feedforward neural network fuzzy grey predictor (FNNFGP). Here, the FNNPID controller is used to drive the system to desired targets. Additionally, a learning mechanism with robust checking conditions is implemented into the FNNPID in order to optimize online its parameters with respect to the control error minimization. Meanwhile, the FGP predictor with self-tuning ability of the predictor step size takes part in, first, estimating the system output in the near future to optimize the controller parameters in advance and, second, creating a compensating control signal accordingly to the system perturbations and, consequently, improving the control performance. Real-time experiments have been carried out to investigate the effectiveness of the proposed control approach.