Optimized Computational Analysis of Feedforward and Feedback Control Scheme using Genetic Algorithm Techniques

A computational analysis using Artificial Intelligence techniques has drawn its attention in Industrial Revolution 4.0. There are self-regulating and integrating processes involved. The commonly used control method involves single-loop feedback control. Embedding feedforward algorithm is aimed to improve the steady state response of the controlled process without affecting the performance of transient response. Apart of it, two optimal tunings are respectively applied to suit to the different control objectives. Correlation PID tunings are applied to respective objective somehow is very relying to the engineering experience and skills. This paper proposed using Genetic Algorithm for optimization analysis to search a trade-off optimized PI controller settings, which reduces dependency on skills as well as provide better insights to all practical engineers for commending effective PI tunings to the real control practices of plants. Initially, the respective process and disturbance models of LOOP-PRO software for both Jacketed-reactor and Pumped-tank were developed. Then, the simulation analysis of PI tunings was conducted to feedback control, feedforward and feedback and Genetic Algorithm scheme. Relative performance was compared in terms of graphs, performance index, and performance indicator. It is concluded that Genetic Algorithm has consistently provided a trade-off optimized PI controller settings for both Jacketed-reactor and Pumped-tank.