Abstract
Traditional approaches to modeling real chemical engineering processes are based on fundamental chemical and physical laws, which include nonlinear algebraic and differential equations. From a computational point of view, these equations have some difficulties with regard to the numerical methods used for their approximation, as well as with the achievement of the desired accuracy of the calculations.
In recent years, there has been a growing interest in the application of the Artificial Neural Networks (ANNs) method to solve a number of problems in the field of chemical engineering related to fault detection, signal processing, modeling, and control of chemical and biochemical processes in which traditional modeling methods have difficulty and it is even impossible to develop physical models with acceptable errors.
Their main advantage is that they work only with data on the input and output values of the process parameters. One model can be used to generate multiple outputs. Once the neural network model is adequately trained and validated, it is able to make predictions for new data about the input values of process parameters that were not used in the development of this model.
This chapter presents the main characteristics of ANNs, the choice of architecture, the process of training and validation of ANN models, as well as several types of ANNs, such as feed-forward nets, recurrent nets and radial basis function nets and combined models with examples of applications in chemical engineering.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Abbreviations
- b :
-
Vectors with values of biased neurons for the hidden and output layer
- B :
-
A matrix of weighting coefficients represented by connections from an input layer to a hidden layer
- c:
-
A vector that centers each RBFN function in the input space
- C :
-
A matrix of weighting coefficients represented by connections links from a hidden layer to an output layer
- D :
-
A matrix of weighting coefficients for the recurrent connections from the output layer at a time t − 1 to the inputs of the hidden layer at a time t
- E :
-
Error of estimation between measured and calculated values of the outputs
- f(⋅):
-
Nonlinear function
- g :
-
Transfer function or activation function
- h :
-
A general signal in which the weighted signals are summed up to a given neuron
- I :
-
Neural network inputs
- J :
-
Standard least squares function
- N :
-
Number of data samples
- O :
-
Neural network outputs
- u :
-
Vector with input values
- v :
-
Common input of each neuron
- w :
-
Weighting coefficients of the connections between input to the hidden layer and hidden layer to the output layer
- x :
-
Vector with values of the hidden layers
- y :
-
Vector with measured values of the outputs
- \( \hat{y} \) :
-
Vector with the values of the outputs predicted (calculated) by the model
- z :
-
A time shift
- Z :
-
A set of the values of the inputs and outputs
- ε :
-
Vector of model error prediction
- \( {\hat{\Phi}}_{\Psi_1{\Psi}_2} \) :
-
Normalized cross-correlation between two variables (time series) Ψ1 and Ψ2
- φ(⋅):
-
Activation functions (transfer functions) of neurons in the hidden and output layers
- θ :
-
Vector with parameters of nonlinear IRN model
- σ :
-
“span” parameter in radial basis function network
- ζ :
-
Positive scalar function
- h :
-
For hidden layer
- i :
-
For neurons (nodes)
- j :
-
For neural network inputs
- o :
-
For the output layer
- t :
-
For time
- ANN:
-
Artificial neural network
- ERN:
-
Externally recurrent network
- IERN:
-
Internal-External recurrent network
- IRN:
-
Internally recurrent network
- MAPE:
-
Mean absolute percentage error
- R:
-
Linear correlation coefficient
- RBFN:
-
Radial basis function network
- RMSE:
-
Root mean square error
- RNN:
-
Recurrent neural network
References
A.E. Rodrigues, М. Minceva, Modelling and simulation in chemical engineering: Tools for process innovation. Comput. Chem. Eng. 29(6), 1167–1183 (2005)
J.D. Hoffman, S. Frankel, Numerical Methods for Engineers and Scientists (Marcel Dekker, New York, 2001)
M. Shacham, N. Brauner, Preventing oscillatory behavior in error control for ODEs. Comput. Chem. Eng. 32(3), 409–419 (2008)
R.E. Precup, S. Preitl, E.M. Petriu, J.K. Tar, M.L. Tomescu, C. Pozna, Generic two-degree-of-freedom linear and fuzzy controllers for integral processes. J. Frank. Instit. 346(10), 980–1003 (2009)
W.J. Cole, K.M. Powell, T.F. Edgar, Optimization and advanced control of thermal energy storage systems. Rev. Chem. Eng. 28(2-3), 81–100 (2012)
Y. Meng, B.L. Lin, A feed-forward artificial neural network for prediction of the aquatic ecotoxicity of alcohol ethoxylate. Ecotoxicol. Environ. Saf. 71(1), 172–186 (2008)
S. Curteanu, Different types of applications performed with different types of neural networks. In: Kwon SJ, editor. Artificial Neural Networks. Chapter 3. New York: Nova Science Publishers, 2011
F.S. Hoseinian, B. Rezai, E. Kowsari, M. Safari, A hybrid neural network/genetic algorithm to predict Zn(II) removal by ion flotation. J. Separat. Sci. Technol. 55(6), 1197–1206 (2020)
S.M. Mousavi, E.S. Mostafavi, P. Jiao, Next generation prediction model for daily solar radiation on horizontal surface using a hybrid neural network and simulated annealing method. Energy Convers. Manag. 153, 671–682 (2017)
M.A. Ahmadi, S. Zendehboudi, A. Lohi, A. Elkamel, I. Chatzis, Reservoir permeability prediction by neural networks combined with hybrid genetic algorithm and particle swarm optimization. Geophys. Prospect. 61(3), 582–598 (2013)
S. Bahrami, F.D. Ardejani, E. Baafi, Application of artificial neural network coupled with genetic algorithm and simulated annealing to solve groundwater inflow problem to an advancing open pit mine. J. Hydrol. 536, 471–484 (2016)
H. Cartwright, Using Artificial Intelligence in Chemistry and Biology (CRC Press, Taylor & Francis Group, UK, 2008)
F.A.N. Fernandes, L.M.F. Lona, Neural network applications in polymerization processes. Braz. J. Chem. Eng. 22(3), 323–330 (2005)
B.M. Wilamowski, Neural network architectures and learning algorithms. IEEE Ind. Electron. Mag. 3(4), 56–63 (2010)
S. Curteanu, Direct and inverse neural network modeling in free radical polymerization. Open Chemistry 2(1), 113–140 (2004)
G. Lisa, W.D. Apreutesei, S. Curteanu, C. Lisa, C.G. Piuleac, V. Bulacovschi, Ferrocene derivatives thermostability prediction using neural networks and genetic algorithms. Thermochim. Acta 521(1-2), 26–36 (2011)
J. Lobato, P. Cañizares, M. Rodrigo, C. Piuleac, S. Curteanu, J. Linares, Direct and inverse neural networks modelling applied to study the influence of the gas diffusion layer properties on PBI-based PEM fuel cells. Int. J. Hydrog. Energy 35, 7889–7897 (2010)
N. Vaklieva-Bancheva, R. Vladova, E. Kirilova, Simulation of heat-integrated autothermal thermophilic aerobic digestion system operating under uncertainties through artificial neural networks. Chem. Eng. Trans. 76, 325–330 (2019a)
E.G. Shopova, N.G. Vaklieva-Bancheva, Basic – A genetic algorithm for engineering problem solution. Comput. Chem. Eng. 30(8), 1293–1309 (2006)
N. Vaklieva-Bancheva, R. Vladova, E. Kirilova, Methodology for energy efficiency and sustainability improvement of batch production systems on the example of autothermal thermophilic aerobic digestion systems. J. Ecol. Eng. 20(9), 103–115 (2019b)
J.L. Elman, Finding structure in time. Cogn. Sci. 14, 179–211 (1990)
F. Caliman, S. Curteanu, C. Betianu, M. Gavrilescu, I. Poulios, Neural networks and genetic algorithms optimization of the photocatalytic degradation of alcian blue 8GX. J. Adv. Oxidat. Technol. 11(2), 316–326 (2008)
E.N. Drăgoi, S. Curteanu, F. Leon, A.I. Galaction, D. Cascaval, Modeling of oxygen mass transfer in the presence of oxygen-vectors using neural networks developed by differential evolution algorithm. Eng. Appl. Artif. Intell. 24(7), 1214–1226 (2011)
E. Kirilova, S. Yankova, B. Ilieva, N. Vaklieva-Bancheva, A new approach for modeling the biotransformation of crude glycerol by using NARX ANN. J. Chem. Technol. Metall. 49(5), 473–478 (2014)
Y. Chen, D.M. Himmelblau. Determination of nonlinear dynamic model order by false nearest neighbor method. In World Congress on Neural Networks, Washington, DC, July 17-21, 1995
J. Lee, C. Yum, W. Kim. Neural network based judgmental adjustment for time series forecasting. in EANN Helsinbu, August 1995, pp. 229
Z. Ahmad, J. Zhang, Selective combination of multiple neural networks for improving model prediction in nonlinear systems modeling through forward selection and backward elimination. Neurocomputing 72, 1198–1204 (2009)
Z-H Zhou, J. Wu, W. Tang. Ensembling neural networks: Many could be better than all. Artif. Intell., vol. 137(1-2), pp. 239-263, 2002
M.H. Nguyen, H.A. Abbass, R.I. McKay, Stop** criteria for ensemble of evolutionary artificial neural networks. Appl. Soft Comput. 6(1), 100–107 (2005)
A. Mukherjee, J. Zhang, A reliable multi-objective control strategy for batch processes based on bootstrap aggregated neural network models. J. Process Control 18(7-8), 720–734 (2008)
F. Herrera, J. Zhang, Optimal control of batch processes using particle swam optimisation with stacked neural network models. Comput. Chem. Eng. 33, 1593–1601 (2009)
J. Torres-Sospedra, C. Hernandez-Espinosa, M. Fernandez-Redondo, Combining MF networks: A comparison among statistical methods and stacked generalization artificial neural networks in pattern recognition, in Springer, ed. by F. Schwenker, S. Marinai, (Lecture Notes Comput Sci, Berlin/Heidelberg, 2006), pp. 210–220
Y. Tian, J. Zhang, A.J. Morris, Modeling and optimal control of a batch polymerization reactor using a hybrid stacked recurrent neural network model. Ind. Eng. Chem. Res. 40, 4525–4535 (2001)
C. Piuleac, M. Rodrigo, P. Cañizares, S. Curteanu, C. Sáez, Ten steps modelling of electrolysis processes by using neural networks. Environ. Model. Softw. 25(1), 74–81 (2010)
F. Leon, C.G. Piuleac, S. Curteanu, Stacked neural network modeling applied to the synthesis of polyacrylamide-based multicomponent hydrogels. Macromol. React. Eng. 4(9-10), 591–698 (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Kirilova, E.G. (2022). Artificial Neural Networks: Applications in Chemical Engineering. In: Boyadjiev, C. (eds) Modeling and Simulation in Chemical Engineering. Heat and Mass Transfer. Springer, Cham. https://doi.org/10.1007/978-3-030-87660-9_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-87660-9_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87659-3
Online ISBN: 978-3-030-87660-9
eBook Packages: Chemistry and Materials ScienceChemistry and Material Science (R0)