Decentralised PID control tuning for a multivariable glass furnace by genetic algorithm

In this paper, standard genetic algorithms (SGAs) are presented to optimise the decentralised discrete PID parameters by three tuning approaches for a multivariable glass furnace with single loop interaction. At initial state, with the aid of control oriented model, the individual tuning approach without loop interaction is considered to categories the genetic operators, personalised cost functions and improved searching boundaries to attain the performance criterions. The second tuning approach is considered in PID parameters optimisation with loop interaction with optimum cost function. While, the third tuning approach is considered by modifying the optimum cost function by including the total effect of glass temperature and excess oxygen of both variables. The modified cost function exhibits better control robustness and disturbance rejection under loop-interaction.