Discrete Time PID Controller on Vortex Induced Vibration for Marine Risers Based on Neural Network Identification Model

Abstract

In this paper, PID controller has been employed to suppress vortex induced vibration (VIV) on a marine riser cylindrical pipe in a good way. Neural Network time series (NNTS) have been represented as a nonlinear system identification to describe the dynamic response of the system after extracting the input-output data from previous paper which it collected from experimental setup. NNTS methods comprised three methods: Neural Network (NARX) based on the Nonlinear Auto-Regressive with External (Exogenous) Input, Neural Network (NAR) based on the Nonlinear Auto-Regressive and Nonlinear Input-Output Neural Network. The Discrete Time PID controller (DT- PID) based on heuristic methods has been chosen to attenuate the VIV on the best model of the three methods. Mean square Error (MSE) has been selected to validate from the nonlinear system identification and PID parameters. Finally, the outcomes showed that the NARX is the best method to represent the VIV on pipe riser from other methods which recorded least MSE= 1.2714×10-9 and DT-PID controller succeeded effectively to suppress vortex induced vibration for marine pipe risers under disturbance load.