Improved Data-Driven Networked Predictive Control Methods

In recent years,networked control systems have been widely used in industry,transportation,power,robotics,medical treatment,smart home and other fields due to its many advantages such as simple structure,high flexibility and reliability,easy expansion and maintainability,long-distance distributed control and so on.However,with the introduction of communication network,networked control systems have also many problems,such as packet dropout,network-induced delay and so on.Therefore,in this paper,an improved data-driven networked predictive control schemes are proposed for a class of discrete-time nonlinear systems with random packet dropout and network-induced delay so as to compensate for the adverse effect of these communication constraints actively.Furthermore,the comparisons between the improved methods and the original methods are given.The main work is as follows:1)In order to overcome the deficiency of compact format dynamic linearization based model-free adaptive control(CFDL-MFAC)method,an improved MFAC method is proposed and its stability is analyzed,which verifies the feasibility of the improved method theoretically.Meanwhile,the control performance of the improved method ard the original method is compared by simulation and experiment.The results of simulation and experiment show that the control performance of the improved method is better than that of the original CFDL-MFAC.2)For a class of discrete-time networked nonlinear systems with random packet dropout in the feedback and forward channels,an improved data-driven network predictive control method is proposed based on the improved MFAC method above.The compensation effect of the improved data-driven network predictive control method and the original method is studied by simulation and experiment.The results of simulation and experiment verify.the effectiveness of the improved method.3)For a class of discrete-time networked nonlinear systems with random packet dropout and network-induced delay in feedback and forward channels,the random packet dropout and network-induced delay are transformed into network delay by a unified way.An improved data-driven network predictive control method is proposed based on the improved MFAC method above.The compensation effect of the improved data-driven network predictive control method and the corresponding original method is studied by simulation and experiment.The results of simulation and experiment prove the effectiveness of the improved scheme.