A New Approach to the Identification of Distillation Column Based on Hammerstein Model

Abstract

Developing a suitable nonlinear model is the most challenging problem in the application of nonlinear model based controllers to distillation column. Hammerstein model consists of a nonlinear static element described by wavenet based nonlinear function, followed by a linear dynamic element described by the Output Error(OE) model was used in this study to represent the nonlinear dynamics of the distillation column. The model parameters were identified using iterative prediction-error minimization method. The model validation results proved that the Hammerstein model was capable of capturing the nonlinear dynamics of distillation column.

Hybrid Series/Parallel All-Nonlinear Dynamic-Static Neural Networks: Development, Training, and Application to Chemical Processes

“…The Hammerstein type of structure consists of a nonlinear static model followed by a linear dynamic model, whereas the Wiener type is basically the reverse, i.e., a linear dynamic model followed by a nonlinear static model. The Hammerstein, Wiener, Hammerstein–Wiener, and Wiener–Hammerstein type of models have been widely used for representing many chemical process systems such as a continuous stirred tank reactor (CSTR) system, a pH neutralization process, a polymerization reactor, distillation towers, , etc. In those studies, static nonlinearities are represented by different types of nonlinear models such as polynomial functions, ANNs, wavelets, etc.…”

“…The Hammerstein, Wiener, Hammerstein–Wiener, and Wiener–Hammerstein type of models have been widely used for representing many chemical process systems such as a continuous stirred tank reactor (CSTR) system, a pH neutralization process, a polymerization reactor, distillation towers, , etc. In those studies, static nonlinearities are represented by different types of nonlinear models such as polynomial functions, ANNs, wavelets, etc. while the linear time invariant (LTI) dynamic model is typically represented by the state space model, , transfer functions, , Laguerre models, finite impulse response (FIR) models, etc.…”

A New Identification Approach of MIMO Hammerstein Model with Separate Nonlinearities

“…Many process (chemical and biological process, signal processing, etc.) have this structure, for example: pH neutralization processes [2], polymerization reactor [3], distillation columns [4] and dryer process [5]. Many system identification methods have been used to identify the single-input single-output (SISO) model.…”

Controle Preditivo Baseado Em Modelo Não Linear Aplicado a Uma Coluna Despropanizadora

“…Os tipos de modelos não-lineares mais utilizados para NMPC são Hammerstein-Winer. O Modelo que mostrou melhor ajuste e menor perda de função foi o Hammerstein-Wiener Kanthasamy et al (2014). aplicou um NMPC com modelo de Hammerstein com bloco estacionário não-linear modelado por wavenet e bloco linear dinâmico por erro de saída a uma coluna de destilação.…”