Fractional Dual-Tilt Control Scheme for Integrating Time Delay Processes: Studied on a Two-Tank Level System
Dipjyoti Das, Sudipta Chakraborty, Utkal Mehta, G. Lloyds Raja
Abstract
Industrial processes of time delayed integrating type require sophisticated control methods because of their non-self-regulating nature. Literature presents fractional order controllers as possible solution to this problem and there is scope for further enhancing the performance of the existing solutions. Though fractional order controllers are capable of outperforming their integer order counterparts, it is evident in literature that the former controller designs are complex and hence analytical tuning procedures for the same are scarce. Hence, this paper presents a novel predictive strategy for time-delayed integrating processes based on two fractional-order tilt-derivative (FOT <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">α</sup> D <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1−α</sup> ) controllers. These controllers are analytically designed with only two tunable parameters. These tuneables are computed using gain- and phase-margin specifications. In contrast to the previous methods, the presented scheme is more capable of eliminating input-load disturbances without adding complexity in terms of tunable parameters. In addition to the investigations carried out using three difficult benchmark plant models, the present design is also experimentally validated using a two-tank level control system to vindicate its efficacy. Through robust stability analysis, it is shown that the suggested strategy is capable of achieving stable closed-loop responses amid upto 50% perturbation in plant parameters.