ISSN 2630-0583 (Print)
ISSN 2630-0656 (Online)
Journal of Current Science and Technology
Rangsit Journal of Arts and Sciences. Vol.6 No.2 , July - December 2016.
Time-delay estimator and disturbance observer based on neural network in networked control system
In this paper, we deal with the control problem of network induced delays and randomly varying time-delay controlled plant under the effect of disturbances and noise in networked control systems (NCS). In a time-delay NCS it becomes more challenging to attain stability when the disturbances and noise interference appear in form of a time-varying signal in the close loop of the NCS. These in turn make the conventional control methods, e.g., normal mathematical model of Smith predictor, more complicated when the aim is to meet quality requirements of the NCS. To overcome these inherent challenges, we mainly analyze the existing techniques, and then propose a novel method to ef?ciently reduce the effect of time-delays, disturbances, and noise interference for highly ef?cient and accurate control purposes. Speci?cally, we introduce a joint solution of time-delay estimator and disturbance observer in which the outer loop with an adaptive Smith predictor is utilized to compensate time-delays for the whole NCS while the inner loop with disturbance observer is to eliminate the disturbances and noise interference. By using neural network identi?cation and the estimation method, the proposed model provides many outstanding advantages such as high adaptation, robust stability, and fast response. The simulation results generated via TrueTime Beta2.0 platform demonstrate that our design signi?cantly improves the performance of NCS.