Abstract:Highlights • GFNN and MFNN models are presented for chemical processes modelling. • MFNN performs better in modelling chemical processes. • A cycle-avoidance approach is presented for cycle processes modelling by MFNN. • MFNN model has a partial interpretability and physical meaning. Background Chemical process modeling is essential for the optimization, precise control and prediction of processes performance. As the modeling based on the first principles is time-consuming and labor-intensive, utilizing neural networks for the process modeling was proposed. However, the poor interpretability of the neural network models is may a problem in the modelling of chemical processes where the accuracy and reliability of the model are highly required. Method Inspired by the functional partitioning of neural networks in human brain, in this paper, global fully connected neural network (GFNN) and modular fully connected neural network (MFNN) models are constructed for the production process of ethylbenzene and that of isopropanol with cycle process, respectively. For the chemical processes with complex cycle processes, the quality of the product can be easily affected by small feed changes due to the highly nonlinear model which is called snowball effect. To avoid snowball effect, the cycle-avoidance MFNN is constructed, and it shows better modelling performance than GFNN. Significant findings MFNN model has more advantages than GFNN model in the modelling of large-scale chemical processes without cycle processes. In addition, it is clear that MFNN model has a partial interpretability which means the local physical meaning for different parts of the whole process. Graphical abstract Download : Download high-res image (251KB) Download : Download full-size image

Non-linear Chemical Process Modelling and Application in Epichlorhydrine Production Plant Using Wavelet Networks

Non-linear Model Predictive Control Based on Neural Network Model with Modified Differential Evolution Adapting Weights

Genetic Algorithm Based Recurrent Fuzzy Neural Network Modeling of Chemical Processes

Application study of a kind of novel neural networks models

Recurrent Neural Network Integrated with Process-Priori-knowledge and Its Application

Wavelet Neural Network for Modeling Chlorophyll a Concentration Affected by Artificial Upwelling

Study on WN Applied in Ferment Process

Control and Optimization of EPI Reaction Ristillation Process Based on Wavelet Networks

Application Research of Wavelet Neural Networks in Process Control

Application of Combining Prior Knowledge and Neural Network in Biochemical Process Modeling

Adaptive nonlinear model predictive control for a class of multivariable chemical processes

Nonlinear Semi-Parametric Modeling Mothed Based on GA-ANN

Dynamic Modeling And Nonlinear Predictive Control Based On Partitioned Model And Nonlinear Optimization

Nonlinear semi-parametric modeling method based on GA-ANN

Modeling and Control of Processes with Output Dynamic Nonlinearity

Nonlinear predictive control based on wavelet neural network applied to polypropylene process

Application of Artificial Neural Networks in Chemical Process Control

Wavelet Networks Based Soft Sensors And Predictive Control In Fermentation Process

Process structure-based fully connected neural network for the modelling of chemical processes: A comparison between global and modular configurations

The Application of Wavelet Neural Networks to Nonlinear Predictive Control

Modeling Nonlinear Dynamic Chemical Process Base on Artificial Neural Networks