Application of an improved augmented Lagrangian algorithm to the tuning of robust PID controller for hydraulic turbine governing system

Abstract

The tuning of a proportional-integral-derivative (PID) controller for hydraulic turbine governing system is a constrained engineering problem. The PID control parameters must be adjusted to capture the requirements of achieving satisfactory transient behaviour, robust stability, and disturbance rejection, which are all essential aspects for hydraulic turbine control. An optimal tuning method for robust PID controller based on the augmented Lagrangian (AL) algorithm and particle swarm optimisation (PSO) is presented in this paper. The AL is extended by incorporating the PSO in a seamless manner. Moreover, it introduces a new updated strategy for the penalty factor and embeds a local search procedure within the algorithm framework to enhance the search capability and accuracy, so that the optimal control parameters can be reliably obtained from the search space. The results of the simulation experiment demonstrate the effectiveness of this improved AL algorithm in producing a robust PID controller that provides system stability over a wide range of operating conditions.