Event‐based neural network predictive controller application for a distillation column

Abstract

In this work, the case study is a distillation column, which is a multi‐input multi‐output (MIMO) nonlinear process. An event‐based neural network predictive controller is utilized for the case study, considering control and energy policies. Computation and communication reduction are the main purposes of the event‐based strategy. The event‐based model predictive controller also copes successfully with the multi‐input multi‐output (MIMO) time‐delayed nonlinear processes. In order to achieve a suitable nonlinear data‐based model of the process, an event‐based neural network predictive controller is proposed. Moreover, new Cuckoo Optimization Algorithm (COA) is employed to improve the efficiency of the neural network. Evaluation of the proposed controller has illustrated a satisfactory performance in both setpoint tracking and disturbance rejection while computational load has been significantly decreased.

An algebraic and suboptimal solution of constrained model predictive control via tangent hyperbolic function

“…Theorem 1. The control law f(x 0 (t)) and (5) with N ≥ 1 is persistently feasible if X f is a control invariant set for the system (1).…”

“…In this example, we want to observe the control performance under nonlinearity conditions by holding the challenging sampling time. Constraints vectors are, x c = [2 700 50000] T and u c = [5]; the initial condition is x 0 = 0.5x c ; weighting coefficients are β 2 1 = 1 , ρ = 1 ∂ 2 1 = 0:9, ∂ 2 2 = 0:09,∂ 2 3 = 0:01.…”

“…Due to the challenges in finding the optimal solution for long sampling times that are primarily required for the control of the systems with fast dynamics, the control equipment requires high-speed processors, which increase the installation costs. Therefore, the industrial applications of MPC are often used in the low-order and relatively "slow" processes [5,6]. Hence, many researchers focus on increasing the speed of the solution to improve the applicability of MPC [7,8].…”

Experimental performances of nonlinear GPC and discrete–time PID in multivariate process: NARIMAX and ARX models in temperature control of reactive distillation column in synthesis of biodiesel from waste cooking oil

“…Giwa et al [16] implemented MATLAB/Simulink's MPC toolbox for renewable energy. Regarding the multivariable theoretical control, Wu [17] proposed an improved PID controller optimized with an extended non-minimal state space model based MPC, Saravanakumar et al [18] used Lagrange-based state transition algorithm to tune the decentralized PID controller, using its numerical stability and performance, Abraham et al [19] developed an optimal GPC using the first principle and linearized 16 th order and reduced fifth order models, Hadian et al [20] used an event-based neural network prediction controller using the Cuckoo Optimization Algorithm for the nonlinear process, Shin et al [21] used HYSYS and examined an MPC integrated neural network model in the nonlinear process, and Cheng et al [22] proposed a dynamic decoupling strategy based on active disturbance rejection control for a first order system with an observer based on a nonlinear wave model.…”

Speed control of permanent magnet synchronous motor using neural network model predictive control

“…The neural network model predictive control does not require any mathematical model of the system, plant model is updated in offline mode by using the collected data from the plant and it reduces the online computation burden [10][11][12]. The neural network model predictive control technique has two processes, the first process is plant identification and the second process is predictive control mechanism using receding horizon policy [13][14][15].…”