DOI QR코드

DOI QR Code

Predicting flux of forward osmosis membrane module using deep learning

딥러닝을 이용한 정삼투 막모듈의 플럭스 예측

  • Kim, Jaeyoon (Department of Civil Engineering, Pukyong National University) ;
  • Jeon, Jongmin (Department of Civil Engineering, Pukyong National University) ;
  • Kim, Noori (Department of Civil Engineering, Pukyong National University) ;
  • Kim, Suhan (Department of Civil Engineering, Pukyong National University)
  • Received : 2020.12.29
  • Accepted : 2021.01.29
  • Published : 2021.02.15

Abstract

Forward osmosis (FO) process is a chemical potential driven process, where highly concentrated draw solution (DS) is used to take water through semi-permeable membrane from feed solution (FS) with lower concentration. Recently, commercial FO membrane modules have been developed so that full-scale FO process can be applied to seawater desalination or water reuse. In order to design a real-scale FO plant, the performance prediction of FO membrane modules installed in the plant is essential. Especially, the flux prediction is the most important task because the amount of diluted draw solution and concentrate solution flowing out of FO modules can be expected from the flux. Through a previous study, a theoretical based FO module model to predict flux was developed. However it needs an intensive numerical calculation work and a fitting process to reflect a complex module geometry. The idea of this work is to introduce deep learning to predict flux of FO membrane modules using 116 experimental data set, which include six input variables (flow rate, pressure, and ion concentration of DS and FS) and one output variable (flux). The procedure of optimizing a deep learning model to minimize prediction error and overfitting problem was developed and tested. The optimized deep learning model (error of 3.87%) was found to predict flux better than the theoretical based FO module model (error of 10.13%) in the data set which were not used in machine learning.

Keywords

References

  1. Bae, K.T., Kim, C.J. (2016). An agricultural estimate price model of artificial neural network by optimizing hidden layer, J. Korean Inst. Inform. Technol., 14(12), 161-169. https://doi.org/10.14801/jkiit.2016.14.12.161
  2. Caruana, R., Lawrence, S., and Giles, L. (2002). Overfitting in neural nets: backpropagation, conjugate gradient, and early stopping, Adv. Neural. Inf. Process. Syst., 402-408.
  3. Cho, Y.H., Seo, Y.D., Park, D.J., and Jeong, J.C. (2016). Study on the activation functions for efficient learning in DNN, J. Inst. Electron. Inform. Eng., 800-803.
  4. Ciresan, D.C., and Giusti, A. (2012). Deep neural networks segment neuronal membranes in electron microscopy images, Adv. Neural. Inf. Process. Syst., 12(2), 2843-2851.
  5. Ciresan, D.C., Meier, U., and Masci, J. (2012). Multi-Column Deep Neural Network for Traffic Sign Classification, Neural. Netw., 32, 333-338. https://doi.org/10.1016/j.neunet.2012.02.023
  6. Jang, I.D. and Wee, S.M. (2001). The analysis telecommunication service market with data mining, J. Korea Inform. Sci. Soc., 28(2), 1-3.
  7. Jeon, J., Choi, J.Y., Sohn, J., and Kim, S. (2018a). Performance analysis of a spiral wound forward osmosis membrane module, J. Korea Soc. Environ. Eng., 40(12), 481-486. https://doi.org/10.4491/ksee.2018.40.12.481
  8. Jeon, J., Kim, N., Choi, J.Y., and Kim, S. (2019). Applicability of statistics-based forward osmosis module models, J. Korea Soc. Environ. Eng., 41(11), 611-618. https://doi.org/10.4491/KSEE.2019.41.11.611
  9. Jeon, J., Jung, J., Lee, S., Choi, J.Y., and Kim, S. (2018b). A simple modeling approach for a forward osmosis system with a spiral wound module, Desalination, 433, 120-131. https://doi.org/10.1016/j.desal.2018.01.004
  10. Ji, S., Park, J. (2020). Improvement of existing machine learning methods of digital signal by changing the step-size, J. Digit. Converg., 18(2), 261-268. https://doi.org/10.14400/JDC.2020.18.2.261
  11. Jo, J.M. (2019). Effectiveness of normalization pre-processing of big data to the machine learning performances, J. Inst. Electron. Inform. Sci., 14(3), 547-552.
  12. Joo, G., Park, C., and Im, H. (2020). Performances evaluation of machine learning optimizers, J. Inst. Korean Electr. Electron. Eng., 24(3), 766-776.
  13. Kim, J.E., Blandin, G., Phuntsho, S., Verliefde, A., Le-Clech, P., and Shon, H.K. (2017). Practical considerations for operability of an 8" spiral wound forward osmosis module: Hydrodynamics, fouling behaviour and cleaning strategy, Desalination, 404, 249-258. https://doi.org/10.1016/j.desal.2016.11.004
  14. Kim, J.E., Phuntsho, S., Ali, S.M., Choi, J.Y., and Shon, H.K. (2018). Forward osmosis membrane modular configurations for osmotic dilution of seawater by forward osmosis and reverse osmosis hybrid system, Water Res., 128, 183-192. https://doi.org/10.1016/j.watres.2017.10.042
  15. Kim, S., Paudel, S., and Seo, G.T. (2015). Forward osmosis membrane filtration for microalgae harvesting cultivated in sewage effluent, Environ. Eng. Res., 20, 99-104. https://doi.org/10.4491/eer.2015.005
  16. Kum, D., Ryu, J., Sung, Y., Han, J., and Lim, K.J. (2017). Development and Assessment for extended daily streamflow regression equation of TMDL station using Machine Learning, Korean Soc. Water Environ., 289-290.
  17. Lee, D.Y., and Chang, B.H. (2020a). A study on development of a prediction model for Korean music box office based on deep learning, Int. J. Cotents, 20(8), 10-18.
  18. Lee, J., Choi, J.Y., Choi, J.S., and Kim, S. (2017). A statistics-based forward osmosis membrane characterization method without pressurized reverse osmosis experiment, Desalination, 403, 36-45. https://doi.org/10.1016/j.desal.2016.04.023
  19. Lee, S.M., Park, K.D., and Kim, I.K. (2020b). Comparison of machine learning algorithms for Chl-a prediction in the middle of Nakdong River (focusing on water quality and quantity factors), J. Korean Soc. Water Wastewater, 34(4), 277-288. https://doi.org/10.11001/jksww.2020.34.4.277
  20. Google Tensorflow. https://www.tensorflow.org (December 22, 2020).