Random network delay in model based predictive networked control systems

Networked control systems (NCS) provide many advantages for idustrial development. For larger scale systems and more complex systems usage of NCS’s will become mandatory. However usage of networked control systems introduces time-delay uncertainty in closed-loop system dynamics. These time delays are caused by the time sharing of the communication medium as well as computation time necessary for control algorithms, resulting in destabilization of the system and jeopardizing system stability. In this work a novel networked control system architecture that runs under non-ideal network conditions where packet loss and random time delays occur. previously introduced MBPNCS architecture is expanded by including random delay in the communication network. The delays and data losses caused by the communication network are compensated for using the computational power of the computer nodes of the networked control system. The architecture is independent of the control algorithm and uses a model to predict the plant states into the future to generate corresponding control outputs. This approach enables the system to be controlled in a pre-simulated manner and stability can be maintained even with high packet loss probabilities. In this approach, it has to be assured that the predicted control signals are applied while their prediction conditions are still valid. The proposed model based predictive networked control system architecture is simulated on a DC motor. The overall effect is that stability is maintained although the response of the plant to the reference can be delayed. The system remains stable even when there exists network jitter. The number of predictions that have to be made to keep the system running is also examined and a prediction horizon dependent on rise time of plant is proposed