Tuning of Multivariable PI Controllers by BLT Method for TITO Systems

Nonlinear autoregressive-moving average-L2 (NARMA-L2) controller for multivariable ball mill plant

Mineral processing facilities concern an enormous amount of dynamically complex unit operations (due to nonlinearities), for instance ball mill system. Normally, these processes need multivariable controllers to smooth actions by designing for plant constraints such as deadtimes and dynamics interactions. The present work presents a comparison between a classical PI and nonlinear moving average autoregressive-linearization level 2 (NARMA-L2) controllers based on artificial neural network (ANN) for a ball mill system. The manipulated variables of this plant are the rotation velocity (Vr) and the feeding weight (Wf), while the controlled parameters are the hold up (HU) and the mass fraction under 45 μm (P45). The simulation was built in the MATLAB software (Simulink), comparing the actions of PI and NARMA-L2 controllers in the face of operational changes in specific regions (constraints). The performance of proposed controllers was verified by the integral of absolute error (IAE), integral of squared error (ISE), or the integral of time-weighted absolute error (ITAE). The results of simulation showed the validity of the model obtained and the control technique proposed in this paper, which contributes to studies of multivariate controller designs for ball mills with significant applications. Additionally, this paper brings a first hybrid approach (PI/NARMA-L2) with successful implementation described in the literature.

Robust optimal centralized PI controller for a fluid catalytic cracking unit

Fluidized Catalytic Cracking (FCC) is a complex process that arises due to feed composition, non-linearities, and dynamic mass and heat interactions in its components. FCC is difficult to model and monitor in industries, and one of the key reasons is that they are multivariable processes. Such processes are highly interacting and that makes the process of controlling even more difficult. The interaction between loops can be quantified easily by dRGA. An easy and effective way of controlling multivariable processes is to implement a centralized control system, considering the interactions between measured and manipulated variables. In this study, a centralized control system is designed for the riser section of the FCC unit. The dRGA method is modified to enhance the closed-loop response by formulating an optimization problem and obtaining an optimal controller settings. A rigorous simulation studies show an 826% reduction in ISE values, a 309% reduction in IAE values, and a 262% reduction in ITAE value of T r i s ${T}_{ris}$ from the dRGA method to the modified dRGA method. Further, IAE values for Y l p g ${Y}_{lpg}$ are reduced by 29% from dRGA to modified dRGA method and 34% from synthesis to modified dRGA method.

Fractional Order PID Controller Design for Multivariable Systems using TLBO

The multivariable systems have to control by using multiloop controllers and each closed loop controller has unique characteristics. The successful model structure for design of control system is extremely subject to the accurate choice of the tuning parameters ( K_p, K_i, K_d, \lambda, \mu ) of the controller. The choice of optimal tuning parameters of Fractional Order PID (FOPID) controller leads to accurate controlling of desired level in multivariable system. Here, for multivariable system, a FOPID controller design based on the advanced optimization technique called Teaching Learning based optimization (TLBO) algorithm is proposed. The goal of paper is (i) The elimination of interaction between the control loops and (ii) Reference tracking along the disturbance in each loop. These objectives are satisfied by using four cost function, namely, integral absolute error (IAE), integral square error (ISE), integral time absolute error (ITAE) and integral time square error (ITSE). Out of these cost functions, ITAE based FOPID controller design using TLBO algorithm provides better performance in terms of fast reference tracking and disturbance elimination in the loop. Moreover, the comparative analysis of convergence characteristics of each objective of the controller by using TLBO is presented. The simulation study confirms that the TLBO algorithm based FOPID controller for multivariable systems (2 × 2) are more robust and exhibits superior response with respect to other algorithm.