Model-based recursive networked predictive control

Abstract

A recursive networked predictive control (RNPC) approach is proposed for networked control systems (NCSs), which mainly consists of two parts: a control prediction generator (CPG) and a network delay compensator (NDC). Based on a NARMA model, the CPG is applied to generate control predictions using the historical input-output data of the plant. The NDC is designed in the actuator to actively compensate for the network communication constraints such as network-induced delay, data packet disorder, accumulation and dropout. The RNPC is easy to be implemented in practice compared with previous results in that the recursive method is used to derive the future output predictions and control predictions, and the round-trip time delay is also used in the compensation scheme. Two RNPC systems are designed for a DC motor Internet-based control system, which are based on the nonlinear model and the simply linearized model, respectively. Practical experiments have been carried out to demonstrate the effectiveness of the proposed approach.