Application of generalized predictive control with neural network error correction in networked control system

Abstract

With regards to the characteristics of uncertain delay of networked control system, two control schemes are introduced. To overcome the influence of modeling error and uncertainty of a system, an idea of using generalized predictive control (GPC) is proposed in the networked control system, an error predictive model is further built based on BP neural network, and error predictive value is used to compensate output predictive value, which forms a new control algorithm of GPC. A networked control system based on Ethernet network was constructed, random-delay-time characteristic of data transmission was tested. Three control methods are compared in the system: PID, GPC, and GPC with error correction. GPC with error correction is verified to have the best performance of three methods.