Recently, with the development of the information technology and the expansion of controlproblem, networked control systems have obtained more attention. NCS is closed through areal-time network that are linked by sensors, actuators and controllers. At the same time, theintroducing of communication networks into the control loops makes the analysis and design ofNCS complicated,such as data transmission delay, network lost package, etc.The contents of thispaper is part of fuzzy robust filtering of NSFC(NO:60974071).This paper analyses the background, significance and the present research situation. Inaddition, the paper summarizes networked control systems and the basic problems of NCS. Inview of time delay and lost package problem, the state feedback controller and output feedbackcontroller are designed. It is more serious for the loss of data packets and the effect on thestability to network control system. The purpose of the thesis is mainly to predict the packetloss through joining BP neural network on NCS,to prepare for the compensation of packetloss.The BP neural network predictor is optimized by using particle swarm algorithm.Simulation results show that, compared with the BP neural network, PSO-BP neural networkhas better network packet loss prediction performance. |