Skip to main content

Advertisement

SpringerLink
Log in

Modeling and Optimization of Electrodialytic Desalination of Fish Sauce Using Artificial Neural Networks and Genetic Algorithm

  • Original Paper
  • Published:
Food and Bioprocess Technology Aims and scope Submit manuscript

Abstract

Electrodialysis (ED) has been proposed as a means to reduce sodium ion concentration in fish sauce. However, no information is so far available on the optimum condition to operate the ED process. Artificial neural network (ANN)-based models were therefore developed to predict the ED performance and changes in selected quality attributes of ED-treated fish sauce; optimum operating condition of the process was then determined via multi-objective optimization using genetic algorithm (MOGA). The optimal ANN models were able to predict the ED performance with R 2 = 0.995, fish sauce basic characteristics with R 2 = 0.992, and the concentrations of total aroma compounds and total amino acids, flavor difference, and saltiness of the treated fish sauce with R 2 = 0.999. Through the use of MOGA, the optimum condition of the ED process was the use of an applied voltage of 6.3 V and the maintenance of the residual salt concentration of the treated fish sauce of 14.3 % (w/w).

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  • Baker, R. W. (2004). Membrane technology and applications (2nd edition). West Sussex, UK: Wiley.

    Book  Google Scholar 

  • Belitz, H. D., Grosch, W., & Schieberle, P. (2009). Food chemistry (4th edition). Berlin, Germany: Springer.

    Google Scholar 

  • Boyaci, İ. H., Sumnu, G., & Sakiyan, O. (2009). Estimation of dielectric properties of cakes based on porosity, moisture content, and formulations using statistical methods and artificial neural networks. Food and Bioprocess Technology, 2, 353–360.

    Article  CAS  Google Scholar 

  • Chen, C. R., & Ramaswamy, H. S. (2002). Modeling and optimization of variable retort temperature (VRT) thermal processing using coupled neural networks and genetic algorithms. Journal of Food Engineering, 53, 209–220.

    Article  Google Scholar 

  • Chen, C. R., Ramaswamy, H. S., & Alli, I. (2001). Prediction of quality changes during osmo-convective drying of blueberries using neural network models for process optimization. Drying Technology, 19, 507–523.

    Article  Google Scholar 

  • Chindapan, N., Devahastin, S., & Chiewchan, N. (2009). Electrodialysis desalination of fish sauce: electrodialysis performance and product quality. Journal of Food Science, 74, E363–E371.

    Article  CAS  Google Scholar 

  • Chindapan, N., Devahastin, S., Chiewchan, N., & Sablani, S. S. (2011). Desalination of fish sauce by electrodialysis: effect on selected aroma compounds and amino acid compositions. Journal of Food Science, 76, S451–S457.

    Article  CAS  Google Scholar 

  • Choi, E.-Y., Choi, J.-H., & Moon, S.-H. (2002). An electrodialysis model for determination of the optimal current density. Desalination, 153, 399–404.

    Article  Google Scholar 

  • Fathi, M., Mohebbi, M., & Razavi, S. M. A. (2011). Effect of osmotic dehydration and air drying on physicochemical properties of dried kiwifruit and modeling of dehydration process using neural network and genetic algorithm. Food and Bioprocess Technology, 4, 1519–1526.

    Article  Google Scholar 

  • Fazzalari, F. A. (1978). Compilation of odor and taste threshold values data. West Conshohocken, PA: ASTM Data Series DS 48A.

    Google Scholar 

  • Giri, A., Osako, K., Okamoto, A., & Ohshima, T. (2010). Olfactometric characterization of aroma active compounds in fermented fish paste in comparison with fish sauce, fermented soy paste and sauce products. Food Research International, 43, 1027–1040.

    Article  CAS  Google Scholar 

  • Goncalves, E. C., Minim, L. A., Coimbra, J. S. R., & Minim, V. P. R. (2005). Modeling sterilization process of canned foods using artificial neural networks. Chemical Engineering and Processing, 44, 1269–1276.

    Article  CAS  Google Scholar 

  • Islam, Md. R., Sablani, S. S., & Mujumdar, A. S. (2003). An artificial neural network model for prediction of drying rates. Drying Technology, 21, 1867–1884.

    Article  Google Scholar 

  • Izadifar, M., & Jahromi, M. Z. (2007). Application of genetic algorithm for optimization of vegetable oil hydrogenation process. Journal of Food Engineering, 78, 1–8.

    Article  CAS  Google Scholar 

  • Jiang, J.-J., Zeng, Q.-X., & Zhu, Z.-W. (2011). Analysis of volatile compounds in traditional Chinese fish sauce (Yu lu). Food and Bioprocess Technology, 4, 266–271.

    Article  CAS  Google Scholar 

  • Kerdpiboon, S., Kerr, W. L., & Devahastin, S. (2006). Neural network prediction of physical property changes of dried carrot as a function of fractal dimension and moisture content. Food Research International, 39, 1110–1118.

    Article  Google Scholar 

  • Konak, A., Coit, D. W., & Smith, A. E. (2006). Multi-objective optimization using genetic algorithms: a tutorial. Reliability Engineering & System Safety, 91, 992–1007.

    Article  Google Scholar 

  • Mohebbi, M., Fathi, M., & Shahidi, F. (2011). Genetic algorithm-artificial neural network modeling of moisture and oil content of pretreated fried mushroom. Food and Bioprocess Technology, 4, 603–609.

    Article  CAS  Google Scholar 

  • Sablani, S. S., Baik, O.-D., & Marcotte, M. (2002). Neural networks for predicting thermal conductivity of bakery products. Journal of Food Engineering, 52, 299–304.

    Article  Google Scholar 

  • Sablani, S. S., Ramaswamy, H. S., Sreekanth, S., & Prasher, S. O. (1997). Neural network modeling of heat transfer to liquid particle mixtures in cans subjected to end-over-end processing. Food Research International, 30, 105–116.

    Article  Google Scholar 

  • Shihani, N., Kumbhar, B. K., & Kulshreshtha, M. (2006). Modeling of extrusion process using response surface methodology and artificial neural networks. Journal of Engineering Science and Technology, 1, 31–40.

    Google Scholar 

  • Thai Industrial Standards Institute. (1983). Local fish sauce (pp. 1–30). Bangkok, Thailand: TIS 3-2526.

    Google Scholar 

  • Tungkawachara, S., Park, J. W., & Choi, Y. J. (2003). Biochemical properties and consumer acceptance of Pacific whiting fish sauce. Journal of Food Science, 68, 855–860.

    Article  CAS  Google Scholar 

  • Youssefi, S., Emam-Djomeh, Z., & Mousavi, S. M. (2009). Comparison of artificial neural network (ANN) and response surface methodology (RSM) in the prediction of quality parameters of spray-dried pomegranate juice. Drying Technology, 27, 910–917.

    Article  Google Scholar 

Download references

Acknowledgments

The authors express their sincere appreciation to the Thailand Research Fund (TRF) for supporting the study financially.

Author information

Authors and Affiliations

  1. Department of Food Engineering, Faculty of Engineering, King Mongkut’s University of Technology Thonburi, 126 Pracha u-tid Road, Tungkru, Bangkok, 10140, Thailand

    Nathamol Chindapan, Naphaporn Chiewchan & Sakamon Devahastin

  2. Department of Biological Systems Engineering, Washington State University, Pullman, WA, 99164-6120, USA

    Nathamol Chindapan & Shyam S. Sablani

Authors
  1. Nathamol Chindapan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Shyam S. Sablani
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Naphaporn Chiewchan
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sakamon Devahastin
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Shyam S. Sablani.

Appendix

Appendix

The simple algebraic equations that represent the configurations of three optimal ANN models for the prediction of ED performance and quality of ED-treated fish are listed as follows:

First ANN Model

$$ {x_{{1}}} = {\text{Applied}}\,{\text{voltage}}\left( {\text{V}} \right) $$
$$ {x_{{2}}} = {\text{Residual}}\,{\text{salt}}\,{\text{concentration}}\left( {\% {{\text{w}} \left/ {\text{w}} \right.}} \right) $$
$$ {X_{{1}}} = \left( {0.{25} * {x_{{1}}}} \right) - {1} $$
$$ {X_{{2}}} = \left[ {\left( {{x_{{2}}} - {2}.{65}} \right) * 0.0{9}} \right] - {1} $$
$$ {X_{{3}}} = { \tanh }\left[ {\left( { - {5}.{26}} \right) + \left( {{2}.{49}} \right) * {X_{{1}}} + \left( { - {2}.{72}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{4}}} = { \tanh }\left[ {\left( {0.{73}} \right) + \left( { - 0.{36}} \right) * {X_{{1}}} + \left( { - 0.{48}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{5}}} = { \tanh }\left[ {\left( {0.{9}0} \right) + \left( {{3}.{79}} \right) * {X_{{1}}} + \left( { - 0.{3}0} \right) * {X_{{2}}}} \right] $$
$$ {X_{{6}}} = { \tanh }\left[ {\left( { - {1}.{77}} \right) + \left( {{2}.0{6}} \right) * {X_{{1}}} + \left( {0.{34}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{7}}} = { \tanh }\left[ {\left( {{1}.{37}} \right) + \left( {{3}.{45}} \right) * {X_{{1}}} + \left( {{3}.0{7}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{8}}} = { \tanh }\left[ {\left( {{2}.{68}} \right) + \left( {0.{45}} \right) * {X_{{1}}} + \left( { - {3}.{35}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{9}}} = { \tanh }\left[ {\left( {{2}.{58}} \right) + \left( { - 0.{78}} \right) * {X_{{1}}} + \left( {{1}.{91}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{10}}} = { \tanh }\left[ {\left( { - 0.{52}} \right) + \left( {{2}.{33}} \right) * {X_3} + \left( {{2}.{87}} \right) * {X_4} + \left( {0.{81}} \right) * {X_5} + \left( {0.{77}} \right) * {X_6} + \left( {0.{11}} \right) * {X_7} + \left( {0.{59}} \right) * {X_8} + \left( { - 0.0{3}} \right) * {X_9}} \right] $$
$$ {X_{{{11}}}} = { \tanh }\left[ {\left( {{1}.{23}} \right) + \left( {0.{28}} \right) * {X_{{3}}} + \left( { - {2}.{45}} \right) * {X_{{4}}} + \left( { - {1}.{46}} \right) * {X_{{5}}} + \left( { - {1}.0{2}} \right) * {X_{{6}}} + \left( {0.{22}} \right) * {X_{{7}}} + \left( { - 0.{97}} \right) * {X_{{8}}} + \left( {{1}.0{3}} \right) * {X_{{9}}}} \right] $$
$$ {X_{{{12}}}} = { \tanh }\left[ {\left( {0.{7}0} \right) + \left( { - 0.{7}0} \right) * {X_{{3}}} + \left( { - {1}.{56}} \right) * {X_{{4}}} + \left( { - {1}.{16}} \right) * {X_{{5}}} + \left( { - 0.{63}} \right) * {X_{{6}}} + \left( {0.{7}0} \right) * {X_{{7}}} + \left( { - {1}.{8}0} \right) * {X_{{8}}} + \left( {{2}.{26}} \right) * {X_{{9}}}} \right] $$
$$ {{\text{X}}_{{{13}}}} = { \tanh }\left[ {\left( { - {1}.{46}} \right) + \left( { - 0.{29}} \right)*{{\text{X}}_{{3}}} + \left( {{1}.{89}} \right) * {{\text{X}}_{{4}}} + \left( {{1}.{2}0} \right) * {{\text{X}}_{{5}}} + \left( {0.0{7}} \right) * {{\text{X}}_{{6}}} + \left( { - {1}.{77}} \right) * {{\text{X}}_{{7}}} + \left( {0.{89}} \right) * {{\text{X}}_{{8}}} + \left( { - 0.{33}} \right) * {{\text{X}}_{{9}}}} \right] $$
$$ {\text{Energy}}\,{\text{consumption}} = \left( {{X_{{{1}0}}} + {1}} \right) * \left( {{1}0{7}} \right) $$
$$ {\text{Current}}\,{\text{efficiency}} = {{{\left[ {\left( {{52}.{92} * {X_{{{11}}}}} \right) + {147}.0{8}} \right]}} \left/ {{2}} \right.} $$
$$ {\text{Yield}} = {{{\left[ {\left( {{X_{{{12}}}} + {1}} \right) * \left( {{55}.{3}0} \right) + {89}.{4}} \right]}} \left/ {{2}} \right.} $$
$$ {\text{ED}}\,{\text{Time}} = \left( {{X_{{{13}}}} + {1}} \right) * \left( {{3}0{8}.{75}} \right) $$

Second ANN Model

$$ {x_{{1}}} = {\text{Applied}}\,{\text{voltage}}\left( {\text{V}} \right) $$
$$ {x_{{2}}} = {\text{Residual}}\,{\text{salt}}\,{\text{concentration}}\left( {\% {{\text{w}} \left/ {\text{w}} \right.}} \right) $$
$$ {X_{{1}}} = \left( {0.{25} * {x_{{1}}}} \right) - {1} $$
$$ {X_{{2}}} = \left[ {\left( {{x_{{2}}} - {2}.{65}} \right) * 0.0{9}} \right] - {1} $$
$$ {X_{{3}}} = { \tanh }\left[ {\left( { - {1}.{9}0} \right) + \left( {{1}.{93}} \right) * {X_{{1}}} + \left( { - {1}.{16}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{4}}} = { \tanh }\left[ {\left( { - {3}.{29}} \right) + \left( {{4}.0{2}} \right) * {X_{{1}}} + \left( { - {1}.{54}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{5}}} = { \tanh }\left[ {\left( { - 0.{15}} \right) + \left( {0.{14}} \right) * {X_{{1}}} + \left( {{1}.{87}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{6}}} = { \tanh }\left[ {\left( {{2}.{83}} \right) + \left( {0.0{2}} \right) * {X_{{1}}} + \left( { - {2}.{22}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{7}}} = { \tanh }\left[ {\left( { - {1}.{91}} \right) + \left( { - 0.{59}} \right) * {X_{{1}}} + \left( { - {2}.{81}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{8}}} = { \tanh }\left[ {\left( {{6}.{15}} \right) + \left( { - {1}.{97}} \right) * {X_{{1}}} + \left( {{4}.{68}} \right) * {X_{{2}}}} \right] $$
$$ {X_{{9}}} = { \tanh }\left[ {\left( { - {2}.{47}} \right) + \left( {{2}.{99}} \right) * {X_{{3}}} + \left( {{1}.{12}} \right) * {X_{{4}}} + \left( { - {1}.{57}} \right) * {X_{{5}}} + \left( {{2}.0{9}} \right) * {X_{{6}}} + \left( { - 0.{4}0} \right) * {X_{{7}}} + \left( { - 0.{4}0} \right) * {X_{{8}}})} \right] $$
$$ {X_{{{1}0}}} = { \tanh }\left[ {\left( { - {1}.{43}} \right) + \left( { - 0.{66}} \right) * {X_{{3}}} + \left( {0.{3}0} \right) * {X_{{4}}} + \left( { - 0.{58}} \right) * {X_{{5}}} + \left( {{3}.{83}} \right) * {X_{{6}}} + \left( {{1}.{26}} \right) * {X_{{7}}} + \left( { - {1}.{31}} \right) * {X_{{8}}})} \right] $$
$$ {X_{{{11}}}} = { \tanh }\left[ {\left( {{2}.0{6}} \right) + \left( { - {1}.{29}} \right) * {X_{{3}}} + \left( {0.{62}} \right) * {X_{{4}}} + \left( { - 0.{94}} \right) * {X_{{5}}} + \left( {{1}.{86}} \right) * {X_{{6}}} + \left( {0.{54}} \right) * {X_{{7}}} + \left( { - {3}.{55}} \right) * {X_{{8}}})} \right] $$
$$ {{\text{X}}_{{{12}}}} = { \tanh }\left[ {\left( { - {2}.{78}} \right) + \left( {0.{26}} \right) * {X_{{3}}} + \left( { - 0.{22}} \right) * {X_{{4}}} + \left( { - 0.{54}} \right) * {X_{{5}}} + \left( {{4}.0{1}} \right) * {X_{{6}}} + \left( {0.{96}} \right) * {X_{{7}}} + \left( { - 0.{11}} \right){X_{{8}}})} \right] $$
$$ {X_{{{13}}}} = { \tanh }\left[ {\left( { - {2}.{12}} \right) + \left( {0.{11}} \right) * {X_{{3}}} + \left( {0.00} \right) * {X_{{4}}} + \left( { - 0.{46}} \right) * {X_{{5}}} + \left( {{3}.{29}} \right) * {X_{{6}}} + \left( {{1}.0{8}} \right) * {X_{{7}}} + \left( { - 0.0{2}} \right) * {X_{{8}}})} \right] $$
$$ {X_{{{14}}}} = { \tanh }\left[ {\left( { - {1}.{98}} \right) + \left( {0.{74}} \right) * {X_{{3}}} + \left( { - 0.{51}} \right) * {X_{{4}}} + \left( { - 0.0{8}} \right) * {X_{{5}}} + \left( {{4}.{89}} \right) * {X_{{6}}} + \left( {{4}.{26}} \right) * {X_{{7}}} + \left( {{1}.{93}} \right) * {X_{{8}}})} \right] $$
$$ {X_{{15}}} = { \tanh }\left[ {\left( {0.{23}} \right) + \left( {{2}.{79}} \right) * {X_{{3}}} + \left( { - {1}.{44}} \right) * {X_{{4}}} + \left( {0.{25}} \right) * {X_{{5}}} + \left( {0.{92}} \right) * {X_{{6}}} + \left( {0.{9}0} \right) * {X_{{7}}} + \left( { - 0.0{3}} \right) * {X_{{8}}})} \right] $$
$$ \% \,{\text{change}}\,{\text{in}}\,{\text{total}}\,{\text{nitrogen}}\,{\text{concentration}} = \left( {{X_{{9}}} + {1}} \right) * \left( {{{{{18}.{78}}} \left/ {{2}} \right.}} \right) $$
$$ \% \,{\text{change}}\,{\text{in}}\,{\text{sodium}}\,{\text{ion}}\,{\text{concentration}} = \left( {{X_{{{1}0}}} + {1}} \right) * \left( {{{{{84}.{66}}} \left/ {{2}} \right.}} \right) $$
$$ \% \,{\text{change}}\,{\text{in}}\,{\text{potassium}}\,{\text{ion}}\,{\text{concentration}} = \left( {{X_{{{11}}}} + {1}} \right) * \left( {{{{{94}.{29}}} \left/ {{2}} \right.}} \right) $$
$$ \% \,{\text{change}}\,{\text{in}}\,{\text{total}}\,{\text{soluble}}\,{\text{solids}} = \left( {{X_{{{12}}}} + {1}} \right) * \left( {{{{{54}.{74}}} \left/ {{2}} \right.}} \right) $$
$$ \% \,{\text{change}}\,{\text{in}}\,{\text{density}} = \left( {{X_{{{13}}}} + {1}} \right) * \left( {{{{{12}.{1}0}} \left/ {{2}} \right.}} \right) $$
$$ \% \,{\text{change}}\,{\text{in}}\,{\text{viscosity}} = \left( {{X_{{{14}}}} + {1}} \right) * \left( {{{{{44}.{78}}} \left/ {{2}} \right.}} \right) $$
$$ {\text{Change}}\,{\text{in}}\,{\text{total}}\,{\text{color}}\left( {{\text{r}}E * } \right) = \left( {{X_{{{15}}}} + {1}} \right) * \left( {{{{{23}.{7}0}} \left/ {{2}} \right.}} \right) $$

Third ANN Model

$$ {x_1} = {\text{Residual}}\,{\text{salt}}\,{\text{concentration}}\left( {\% {{\text{w}} \left/ {\text{w}} \right.}} \right) $$
$$ {X_1} = \left[ {0.0{9} * \left( {{x_{{1}}} - {2}.{8}} \right)} \right] - {1} $$
$$ {X_2} = { \tanh }\left[ {\left( {{4}.{85}} \right) + \left( { - {4}.{93}} \right) * {X_1}} \right] $$
$$ {X_3} = { \tanh }\left[ {\left( {0.{2}0} \right) + \left( { - {3}.0{5}} \right) * {X_1}} \right] $$
$$ {X_4} = { \tanh }\left[ {\left( { - {5}.00} \right) + \left( { - {6}.{1}0} \right) * {X_1}} \right] $$
$$ {Y_5} = { \tanh }\left[ {\left( { - {4}.{13}} \right) + \left( {{5}.{56}} \right) * {X_2} + \left( {0.{47}} \right) * {X_{{3}}} + \left( {{1}.{63}} \right) * {X_{{4}}}} \right] $$
$$ {Y_6} = { \tanh }\left[ {\left( { - {5}.{57}} \right) + \left( {0.{64}} \right) * {X_2} + \left( {0.{91}} \right) * {X_{{3}}} + \left( { - {6}.{65}} \right) * {X_4}} \right] $$
$$ {Y_7} = { \tanh }\left[ {\left( {{3}.{54}} \right) + \left( {{1}.0{5}} \right) * {X_2} + \left( {0.{63}} \right) * {X_3} + \left( {{5}.0{8}} \right) * {X_4}} \right] $$
$$ {Y_8} = { \tanh }\left[ {\left( { - {2}.{15}} \right) + \left( {{3}.0{8}} \right) * {X_2} + \left( {0.{7}0} \right) * {X_3} + \left( {{1}.{25}} \right) * {X_4}} \right] $$
$$ \% \,{\text{change}}\,{\text{in}}\,{\text{total}}\,{\text{aroma}}\,{\text{compounds}}\,{\text{concentration}} = \left( {{Y_{{5}}} + {1}} \right) * {36}.{99} $$
$$ \% {\text{ change in total amino acids concentration}} = \left[ {\left( {{Y_6} + {1}} \right) * {14}.{91}} \right] - {16}.{58} $$
$$ {\text{Change}}\,{\text{in}}\,{\text{flavor}}\,{\text{difference}} = \left( {{Y_7} + {1}} \right) * {3}.{15} $$
$$ {\text{Change}}\,{\text{in}}\,{\text{saltiness}}\,{\text{intensity}} = \left( {{Y_8} + {1}} \right) * {2}.{5} $$

Rights and permissions

Reprints and permissions

About this article

Cite this article

Chindapan, N., Sablani, S.S., Chiewchan, N. et al. Modeling and Optimization of Electrodialytic Desalination of Fish Sauce Using Artificial Neural Networks and Genetic Algorithm. Food Bioprocess Technol 6, 2695–2707 (2013). https://doi.org/10.1007/s11947-012-0914-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11947-012-0914-6

Keywords

Search

Navigation