Design of Multivariable Neural Controllers Using a Classical Approach

Abstract

In the present study, the neural network (NN) based multivariable controllers were designed as a series of single input-single output (SISO) controllers or multi variable SISO (MVSISO) controllers utilizing the classical decoupled process models. Multilayer feed forward networks (FFNN) were used as direct inverse neural network (DINN) controllers, which used the inverse dynamics of the decoupled process. To address the disturbance rejection problems, the IMC based neural control architecture was proposed with suitable choice of filter and disturbance transfer function. Multi input-multi output (MIMO) non-linear processes like interacting tank systems, temperature and level control of a mixing tank with hot and cold input streams & a (2×2) distillation process were considered as case studies for that purpose. Simplified as well as ideally decoupled process as well as disturbance transfer functions was used for neural controller design. DINN/IMC based NN controllers performed effectively well in comparison to conventional P/ PI/IMC based PI controllers for set-point tracking & regulator problems.

Loop Interaction Analysis and Control Structure Selection: Application to a Fluid Catalytic Cracking Unit

“…Although they enjoy the merit of well established control theories, control loop interaction posses the threat to -linear multivariate features which result in complex dynamics 1 . Multiple-Input-Multiple-Output (MIMO) controller design procedures are better suited for addressing FCCU control issues, however, due the intricacies associated with the design, tuning and maintenance of the MIMO structure, decentralized PID control systems are largely preferred in the process industry 2,3 . Such decentralized schemes desirable performance.…”

Isıl İşlem Uygulamalarında Kullanılan Bir Karışım Tankının Bulanık Mantık Esaslı Denetimi

“…İki veya daha fazla sıvının karışımından oluşan endüstriyel sıvı karıştırma süreçleri kontrol ve otomasyon sistemleri ile denetlenmektedir [7]. Karışım tankı çok girişli çok çıkışlı ve doğrusal olmayan dinamik bir sistemdir [6,8].…”

Decoupling Multivariable Processes using Partial Least Squares for Decentralized Control

“…One of the earlier approaches of multivariable control had been the decoupling control to reduce the loop interactions. The decoupler combined multivariable processes were used to create as series of ANN-SISO controllers and tuned independently without influencing the performance of other closed loops [27]. In this paper, PLS based decoupling scheme is developed that avoids much load on closed loop control system by removing pre and post compensators (loadings matrices of input and output variables).…”