Colonial competitive algorithm: A novel approach for PID controller design in MIMO distillation column process

Purpose – This paper aims to describe colonial competitive algorithm (CCA), a novel socio-politically inspired optimization strategy, and how it is used to solve real world engineering problems by applying it to the problem of designing a multivariable proportional-integral-derivative (PID) controller. Unlike other evolutionary optimization algorithms, CCA is inspired from a socio-political process – the competition among imperialists and colonies. In this paper, CCA is used to tune the parameters of a multivariable PID controller for a typical distillation column process. Design/methodology/approach – The controller design objective was to tune the PID controller parameters so that the integral of absolute errors, overshoots and undershoots be minimized. This multi-objective optimization problem is converted to a mono-objective one by adding up all the objective functions in which the absolute integral of errors is emphasized to be reduced as long as the overshoots and undershoots remain acceptable. Findings – Simulation results show that the controller tuning approach, proposed in this paper, can be easily and successfully applied to the problem of designing MIMO controller for control processes. As a result not only was the controlled process able to significantly reduce the coupling effect, but also the response speed was significantly increased. Also a genetic algorithm (GA) and an analytical method are used to design the controller parameters and are compared with CCA. The results showed that CCA had a higher convergence rate than GA, reaching to a better solution. Originality/value – The proposed PID controller tuning approach is interesting for the design of controllers for industrial and chemical processes, e.g. MIMO evaporator plant. Also the proposed evolutionary algorithm, CCA, can be used in diverse areas of optimization problems including, industrial planning, resource allocation, scheduling, decision making, pattern recognition and machine learning.