Tracking control method of networked control systems based on predictive compensation

Abstract

This invention discloses a tracking control method of networked control systems based on predictive compensation. A network predictive controller consists of a predictive generator and a time delay compensator; a control increment sequence outputted by the network predictive generator at time k is as shown in the description; and, the control increment of the time delay compensator at time t is as shown in the description. The method reduces the predictive step length at the cost of introducing a slight prediction error, and improves the compensation effect of the predictive controller to the network time delay; and, compared with the traditional network predictive control, the method is easier to be realized.

Description

technical field [0001] The invention relates to a networked control system, in particular to a tracking control method for a networked control system based on predictive compensation. Background technique [0002] The closed-loop control system formed through the communication network is called networked control system (networked control system, abbreviated as NCSs). NCSs has the advantages of convenient installation and maintenance, high flexibility and easy reconfiguration. In the networked control system, the controlled objects, sensors, controllers and actuators are connected through real-time network to form a closed-loop system. The introduction of the network, on the one hand, overcomes the long-distance dispersion of space, and provides the basis for realizing remote monitoring and control. On the other hand, the network topology, limited network resources, and inevitable delay and packet loss in packet transmission communication all pose challenges to traditional c...

AI Technical Summary

Problems solved by technology

For the unmeasurable state of discrete-time linear time-invariant systems, the existing methods design controller-side observers, estimate the state observations several steps ago according to the delay information provided by the feedback channel, and recursively deduce the current and For the state prediction value of several steps in the future, closed-loop control is realized by fixed-gain state feedback, which can improve system performance, but it requires a dedicated channel, which is difficult to implement and high in cost; another improved network predictive control method combines optimal control ideas to realize predictive closed-loop control system, does not require a dedicated channel, but does not take full advantage of known information on the controller side, increasing the prediction step size

Images