Advertisement

Cluster Computing

, Volume 22, Supplement 3, pp 6371–6381 | Cite as

Preference-driven yield-and-quality optimization for high-sulfur gas sweetening process by extreme learning machine model

  • Jianfeng Shang
  • Xiaohua GuEmail author
  • Liping Yang
  • Haihong Tang
  • Kun Zhang
  • Zhongli Ji
Article
  • 176 Downloads

Abstract

Yield and quality, as the two most important output variables of high-sulfur gas (HSG) sweetening process, are affected by the operating parameters. The HSG sweetening process involves more than ten operating parameters, so the relationship between the parameters and output variables is complex, non-linear and strong coupling. This paper tries to use data mining methods to explore this relationship and apply it to optimize the yield and quality. First, a ten (inputs)-to-three (outputs) model is established by extreme learning machine (ELM). Then, a preference-driven multi-objective optimization algorithm is used to maximize the yield while ensuring that the concentration of carbon dioxide (CO2) and the concentration of hydrogen sulfide (H2S) in the treated gas are close to but not exceeding 3% and 4 ppm respectively. The proposed method is validated in a HSG purification plant in southwest China. A set of 3044 production data is collected and randomly divided into 80 and 20% for training and testing. The results show that the established ELM model is in good agreement with the actual operation data. The maximum deviation of mean square error (MSE), mean absolute error (MAE) and average absolute deviation percent (AAD %) of the predictions in three scenarios are 0.2047, 0.3177 and 7.91% respectively. Moreover, the optimization based on the obtained ELM model is also validated. In particular, the H2S concentration and CO2 concentration in the treated gas are significantly higher than those before optimization, but have not exceeding the limits. Thus, the consumptions of energy and amine solvents decreased, while the yield increased.

Keywords

Yield-and-quality optimization High-sulfur gas sweetening process Extreme learning machine Preference-driven multi-objective optimization Data mining 

Abbreviations

HSG

High-sulfur gas

ELM

Extreme learning machine

MSE

Mean square error

MAE

Mean absolute error

AAD%

Average absolute deviation percent

MDEA

N-methyldiethanolamine

AI

Artificial intelligence

NSGA-II

Non-dominated sorting genetic algorithm-II

NNs

Neural networks

SVM

Support vector machine

SLFNs

Single-hidden layer feedforward neural networks

PSO

Particle swarm optimization

PP

Physical programming

HD

Highly desirable

D

Desirable

T

Tolerable

U

Undesirable

HU

Highly undesirable

List of symbols

\(x_{i} ,t_{i}\)

Arbitrary distinct samples

\(\beta_{i}\)

Output weight of the ith hidden layer node connected with the output neuron

\(\omega_{i}\)

Output weight of the input neuron connected with the ith hidden layer node

\(b_{i}\)

Bias of the ith hidden layer node

\(o_{j}\)

Output value corresponding to the jth input sample

\(H^{\dag }\)

Moore–Penrose generalized inverse of matrix H

\(\hat{y}_{i}\)

Model predictions

\(y_{i}\)

Real production data

\(n\)

Number of production data

\(R_{{{\text{lean}} . {\text{tail}}}}\)

Lean amine flowrate to tail gas absorber (t/h)

\(R_{{{\text{lean}} . 2 {\text{nd}}}}\)

Inlet lean amine flowrate of 2nd absorber (t/h)

\(F_{\text{feedgas}}\)

Feed gas flowrate (kNm3/h)

\(R_{\text{semirich}}\)

Semi-rich amine flowrate (t/h)

\(T_{{{\text{lean}} . {\text{tail}}}}\)

Inlet lean amine temperature of 1st absorber(oC)

\(T_{{{\text{lean}} . 2 {\text{nd}}}}\)

Inlet lean amine temperature of 2nd absorber(oC)

\(P_{\text{flash}}\)

Amine flash drum pressure (MPa)

\(Q_{\text{reboilerA}}\)

Steam flowrate of reboiler A (t/h)

\(Q_{\text{reboilerB}}\)

Steam flowrate of reboiler B (t/h)

\(Q_{\text{preheater}}\)

Steam flowrate of preheater (t/h)

\(C_{H2S}\)

H2S concentration in the treated gas (ppm)

\(C_{CO 2}\)

CO2 concentration in the treated gas (mol%)

\(F_{\text{treatedgas}}\)

Treated gas flowrate (kNm3/h)

Notes

Acknowledgements

This work was supported by the National Science and Technology Major Project of the Ministry of Science and Technology of China [Grant Number 2016ZX05017004]; the Chongqing National Science Foundation [Grant Number cstc2015jcyjBX0089]; the National Natural Science Foundation of China [Grant Number 51404051] and the Research Foundation of Chongqing University of Science and Technology [Grant Number CK2016Z16].

References

  1. 1.
    Banat, F., Younas, O., Didarul, I.: Energy and exergical dissection of a natural gas sweetening plant using methyldiethanol amine (MDEA) solvent. J. Nat. Gas Sci. Eng. 16(1), 1–7 (2014)CrossRefGoogle Scholar
  2. 2.
    Shirazian, S., Marjani, A., Rezakazemi, M.: Separation of CO2, by single and mixed aqueous amine solvents in membrane contactors: fluid flow and mass transfer modeling. Eng. Comput. 28(2), 189–198 (2012)CrossRefGoogle Scholar
  3. 3.
    Alhseinat, E., Pal, P., Keewan, M., Banat, F.: Foaming study combined with physical characterization of aqueous MDEA gas sweetening solutions. J. Nat. Gas Sci. Eng. 17(1), 49–57 (2014)CrossRefGoogle Scholar
  4. 4.
    Qiu, K., Shang, J.F., Ozturk, M., Li, T.F., Chen, S.K., Zhang, L.Y., Gu, X.H.: Studies of methyldiethanolamine process simulation and parameters optimization for high-sulfur gas sweetening. J. Nat. Gas Sci. Eng. 21, 379–385 (2014)CrossRefGoogle Scholar
  5. 5.
    Qiu, K., An, P., Yang, F., et al.: Simulation study on impact of operating coditions on energy consumption in high-sulfur natural gas desulfurization. Acta Pet. Sinica 28(6), 162–169 (2012)Google Scholar
  6. 6.
    Taheri, K., Hasanipanah, M., Golzar, S.B., et al.: A hybrid artificial bee colony algorithm-artificial neural network for forecasting the blast-produced ground vibration. Eng. Comput. 33(3), 689–700 (2017)CrossRefGoogle Scholar
  7. 7.
    Ahmad, F., Lau, K.K., Shariff, A.M., et al.: Process simulation and optimal design of membrane separation System for CO2 capture from natural gas. J. Comput. Chem. 36(10), 119–128 (2012)CrossRefGoogle Scholar
  8. 8.
    Behroozsarand, A., Shaffei, S.: Optimal control of distillation column using non-dominated sorting genetic algorithm II. J. Loss Prev. Process. Ind. 24(1), 25–33 (2011)CrossRefGoogle Scholar
  9. 9.
    Behroozsaranda, A., Zamaniyanb, A.: Multiobjective optimization scheme for industrial synthesis gas sweetening plant in GTL process. J. Nat. Gas Chem. 20(1), 99–109 (2011)CrossRefGoogle Scholar
  10. 10.
    Yu, F., Xu, X.: A short-term load forecasting model of natural gas based on optimized genetic algorithm and improved BP neural network. Appl. Energy 134(1), 102–113 (2014)CrossRefGoogle Scholar
  11. 11.
    Salooki, M.K., Abedini, R., Adib, H., Koolivand, H.: Design of neural network for manipulating gas refinery sweetening regenerator column outputs. Sep. Purif. Technol. 83(1), 1–9 (2011)CrossRefGoogle Scholar
  12. 12.
    You, Z., Yu, J., Zhu, L., et al.: A map reduce based parallel SVM for large-scale predicting protein-protein interactions. Neurocomputing. 145(1), 37–43 (2014)CrossRefGoogle Scholar
  13. 13.
    Adib, H., Sharifi, F., Mehranbod, N., Kazerooni, N.M., Koolivand, M.: Support vector machine based modeling of an industrial natural gas sweetening plant. J. Nat. Gas Sci. Eng. 14(1), 121–131 (2013)CrossRefGoogle Scholar
  14. 14.
    Huang, G., Zhu, Q., Siew, C.: Extreme learning machine: theory and applications. Neurocomputing 70(1), 489–501 (2006)CrossRefGoogle Scholar
  15. 15.
    He, Y., Geng, Z., Zhu, Q.: Positive and negative correlation input attributes oriented subnets based double parallel extreme learning machine (PNIAOS-DPELM) and its application to monitoring chemical processes in steady state. Neurocomputing 165(1), 171–181 (2015)CrossRefGoogle Scholar
  16. 16.
    Huang, G.B., Zhu, Q.Y., Siew, C.K.: Extreme learning machine: a new learning scheme of feedforward neural networks [C]. Proceedings of the IEEE International Joint Conference on Neural Networks 2(1), 985–990 (2004)Google Scholar
  17. 17.
    Cao, J., Lin, Z., Bin, G.: Huang. composite function wavelet neural networks with extreme learning machine. Neurocomputing 73(79), 1405–1416 (2010)CrossRefGoogle Scholar
  18. 18.
    Ilgin, M.A., Gupta, S.M.: Physical programming: a review of the state of the art. Studies Inf. Control 21(4), 349–366 (2012)Google Scholar
  19. 19.
    Deb, K., Pratap, A., Agarwal, S., Meyarivan, T.: A fast and elitist multi-objective genetic algorithm: nSGA-II. IEEE Trans. Evol. Comput. 6(2), 182–197 (2002)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  • Jianfeng Shang
    • 1
    • 2
  • Xiaohua Gu
    • 3
    • 4
    Email author
  • Liping Yang
    • 5
  • Haihong Tang
    • 4
  • Kun Zhang
    • 4
  • Zhongli Ji
    • 1
  1. 1.College of Mechanical and Transportation EngineeringChina University of PetroleumBeijingPeople’s Republic of China
  2. 2.Sinopec Zhongyuan Oilfield Puguang BranchSichuanPeople’s Republic of China
  3. 3.Artificial Intelligence Key Laboratory of Sichuan ProvinceSichuan University of Science and EngineeringZigongPeople’s Republic of China
  4. 4.School of Electrical and Information EngineeringChongqing University of Science and TechnologyChongqingPeople’s Republic of China
  5. 5.Laboratory of Optoelectronic Technology and Systems of the Education Ministry of ChinaChongqing UniversityChongqingPeople’s Republic of China

Personalised recommendations