Intelligent predicting of salt pond’s ion concentration based on support vector regression and neural network

  • Jun Liu
  • Aowen XiaoEmail author
  • Guangyuan Lei
  • Guangfeng Dong
  • Mengting Wu
Smart Data Aggregation Inspired Paradigm & Approaches in IoT Applns


The constant dynamic changes in salt pond make it difficult to achieve accurate prediction of ion concentration. It is of great significance to get the accurate prediction of potassium ion concentration in salt pools for the actual production of potash fertilizer. In this paper, some machine learning methods, such as support vector regression (SVR), AdaBoost regressor, K neighbor regressor, gradient boosting regressor, extra trees regressor and neural network regressor, have been used to build the prediction models. In the experiment, the MSE and R2 of the K+ concentration by using SVR in test data set reach 0.26385 and 0.9414, which are better than other models. Therefore, the SVR model has high research value in the field of salt pool ion concentration prediction.


Ion concentration Support vector regression Neural network AdaBoost Gradient boosting Extra trees Potash fertilizer 



This work was supported by the National Natural Science Foundation of China (61172150, 61803286), the Foundation of Hubei Provincial Key Laboratory of Intelligent Robot (HBIR 201802) and the tenth Graduate Innovation Fund of Wuhan Institute of Technology (CX2018197, CX2018200, CX2018212).

Compliance with ethical standards

Conflict of interest

We declare that we do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted.


  1. 1.
    Darst BC (1991) Development of the potash fertilizer industry. Fertil Res 28(1):103–107Google Scholar
  2. 2.
    Hui LH, Chun W, Shi ZW (2014) In the production of potash salt pond salt distribution curve drawing method automatically. Autom Instrum (01):114–115Google Scholar
  3. 3.
    Yang ZS, Yan-Hua LI, Hai-Tao YE (2012) Research on process control of producing carnallite in salt pan. Ind Miner Process 4:15–16Google Scholar
  4. 4.
    Wang Z, Li M, Shao B, WU G, Wu X (2018) Research of natural evaporation of Eyacuo Salt Lake Brine in Tibet. J Salt Chem Ind 47(03):28–30Google Scholar
  5. 5.
    Yan J, Huang XL (2003) Exploration and utilization of main resources of salt lakes in Qinghai Province. Inorg Chem Ind 35(3):5–7Google Scholar
  6. 6.
    Anitha A, Acharjya DP (2017) Crop suitability prediction in Vellore district using rough set on fuzzy approximation space and neural network. Neural Comput Appl 53:1–18Google Scholar
  7. 7.
    Jayashree LS, Palakkal N, Papageorgiou EI, Papageorgiou K (2015) Application of fuzzy cognitive maps in precision agriculture: a case study on coconut yield management of southern India’s Malabar region. Neural Comput Appl 26(8):1963–1978Google Scholar
  8. 8.
    Miao L, Zhu D, Liu M et al (2018) Cooking carbon with protic salt: nitrogen and sulfur self-doped porous carbon nanosheets for supercapacitors. Chem Eng J 347:S1385894718306880Google Scholar
  9. 9.
    Yuntao WU, Longting H, Hui C et al (2014) HOSVD-based subspace algorithm for multidimensional frequency estimation without pairing parameters. Chin J Electron 23(4):729–734Google Scholar
  10. 10.
    Wang Z, Hu J, Wang SZ et al (2015) Trilateral constrained sparse representation for Kinect depth hole filling. Pattern Recognit Lett 65:95–102Google Scholar
  11. 11.
    Liu H, Yuntao WU (2014) Distributed source localization under anchor position uncertainty. Chin J Electron 23(1):93–96Google Scholar
  12. 12.
    Zhong L, Cai D, Wu Y (2013) OHRank: an algorithm integrating mentality and influence of opinion holder for opinion mining. Chin J Electron 22(4):655–660Google Scholar
  13. 13.
    Peng L, Zhang Y, Zhou H et al (2018) A robust method for estimating image geometry with local structure constraint. IEEE Access 6:20734–20747Google Scholar
  14. 14.
    Yuntao WU, Longting H, Hui C et al (2014) HOSVD-based subspace algorithm for multidimensional frequency estimation without pairing parameters. Chin J Electron 23(4):729–734Google Scholar
  15. 15.
    Dwivedi AK (2018) Performance evaluation of different machine learning techniques for prediction of heart disease. Neural Comput ApplGoogle Scholar
  16. 16.
    Dwivedi AK (2017) Analysis of computational intelligence techniques for diabetes mellitus prediction. Neural Comput Appl 9:1–9Google Scholar
  17. 17.
    Dong JR, Zheng CY, Kan GY, Zhao M, Wen J, Yu J (2015) Applying the ensemble artificial neural network-based hybrid data-driven model to daily total load forecasting. Neural Comput Appl 26(3):603–611Google Scholar
  18. 18.
    Asencio-Cortés G, Martínez-Álvarez F, Troncoso A, Morales-Esteban A (2015) Medium-large earthquake magnitude prediction in Tokyo with artificial neural networks. Neural Comput Appl 28(5):1043–1055Google Scholar
  19. 19.
    Lu T, Xiong Z, Zhang Y, Wang B, Lu T (2017) Robust face super-resolution via locality-constrained low-rank representation. IEEE Access 5(99):13103–13117Google Scholar
  20. 20.
    Yoav F, Schapire RE (1996) Experiments with a new boosting algorithm, machine learning. In: Proceedings of the thirteenth international conference, 148–156Google Scholar
  21. 21.
    Tavallali P, Yazdi M, Khosravi MR (2018) Robust cascaded skin detector based on AdaBoost. Multimedia Tools Appl 4:1–22Google Scholar
  22. 22.
    Yan G, Yu M, Yu Y et al (2016) Real-time vehicle detection using histograms of oriented gradients and AdaBoost classification. Opt-Int J Light Electron Opt 127(19):7941–7951Google Scholar
  23. 23.
    Fernández-Delgado M, Reboreda M, Cernadas E, Barro S (2010) A comparison of several neural networks to predict the execution times in injection molding production for automotive industry. Neural Comput Appl 19(5):741–754Google Scholar
  24. 24.
    Friedman Jerome H (2002) Stochastic gradient boosting. Comput Stat Data Anal 38(4):367–378MathSciNetzbMATHGoogle Scholar
  25. 25.
    Ahmad Z, Xie T, Maheshwari C et al (2018) Machine learning enabled computational screening of inorganic solid electrolytes for suppression of dendrite formation in lithium metal anodes. ACS Central Sci 4(8):996–1006Google Scholar
  26. 26.
    Hazar MA, Odabasioglu N, Ensari T, Kavurucu Y, Sayan OF (2017) Performance analysis and improvement of machine learning algorithms for automatic modulation recognition over rayleigh fading channels. Neural Comput Appl 12:1–10Google Scholar
  27. 27.
    Liu JNK, Hu Y (2013) Application of feature-weighted support vector regression using grey correlation degree to stock price forecasting. Neural Comput Appl 22(1 Supplement):143–152MathSciNetGoogle Scholar
  28. 28.
    Myles AJ, Murray AF, Wallace AR, Barnard J, Smith G (1997) Estimating mlp generalisation ability without a test set using fast, approximate leave-one-out cross-validation. Neural Comput Appl 5(3):134–151Google Scholar
  29. 29.
    Baziar S, Shahripour HB, Tadayoni M, Nabi-Bidhendi M (2016) Prediction of water saturation in a tight gas sandstone reservoir by using four intelligent methods: a comparative study. Neural Comput Appl 30(4):1171–1185Google Scholar
  30. 30.
    Gitoee A, Faridi A, France J (2017) Mathematical models for response to amino acids: estimating the response of broiler chickens to branched-chain amino acids using support vector regression and neural network models. Neural Comput Appl 30(8):2499–2508Google Scholar
  31. 31.
    Cherkassky V, Mulier F (1998) Statistical learning theory. Encycl Sci Learn 41(4):3185zbMATHGoogle Scholar
  32. 32.
    Liu J, Wu M, Wang M et al (2018) Predicting the content of camelina protein using FT-IR spectroscopy coupled with SVM model. Cluster Comput 4:1–6Google Scholar
  33. 33.
    Shao W, Li Y, Diao S et al (2016) Rapid classification of Chinese quince (Chaenomeles speciosa, Nakai) fruit provenance by near-infrared spectroscopy and multivariate calibration. Anal Bioanal Chem 409(1):115–120Google Scholar
  34. 34.
    Khozani ZS, Bonakdari H, Zaji AH (2017) Estimating shear stress in a rectangular channel with rough boundaries using an optimized SVM method. Neural Comput Appl 30(8):2555–2567Google Scholar
  35. 35.
    Saidi L, Ali JB, Bechhoefer E et al (2017) Wind turbine high-speed shaft bearings health prognosis through a spectral Kurtosis-derived indices and SVR. Appl Acoust 120:1–8Google Scholar
  36. 36.
    Guo H, Jiao P (2017) Research of state vector in short-term passengers flow forecasting based on nonparametric regression. J Syst Simul 9:2128–2133Google Scholar
  37. 37.
    Tama BA, Rhee KH (2017) An in-depth experimental study of anomaly detection using gradient boosted machine. Neural Comput Appl 1:1–11Google Scholar
  38. 38.
    Marée Raphaël, Geurts P, Wehenkel L (2006) Biological image classification with random subwindows and extra-trees abstract. English 3(1):75–90Google Scholar
  39. 39.
    Beheshti Z, Firouzi M, Shamsuddin SM, Zibarzani M, Yusop Z (2016) A new rainfall forecasting model using the capso algorithm and an artificial neural network. Neural Comput Appl 27(8):2551–2565Google Scholar
  40. 40.
    Ahmed AMAN (2017) Support vector regression-based model for prediction of behavior stone column parameters in soft clay under highway embankment. Neural Comput Appl 3:1–11Google Scholar
  41. 41.
    Mi-Ying Y, Wei-Hua G, Chun-Hua Y (2011) Prediction research on cobalt ion concentration based on gray correlation and imporved support vector machine. J Instrum 32(05):961–967Google Scholar
  42. 42.
    Meihua L, Weihua G, Chunhua Y, Hongqiu Z (2009) Prediction of cobalt ion concentration in the purification process based on WA and LS-SVM. Comput Meas Control 17(04):652–654Google Scholar
  43. 43.
    Zhang J, Wang S (2016) A fast leave-one-out cross-validation for SVM-like family. Neural Comput Appl 27(6):1717–1730Google Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  • Jun Liu
    • 1
    • 2
  • Aowen Xiao
    • 1
    • 2
    Email author
  • Guangyuan Lei
    • 3
  • Guangfeng Dong
    • 3
  • Mengting Wu
    • 1
    • 2
  1. 1.Hubei Key Laboratory of Intelligent RobotWuhan Institute of TechnologyWuhanChina
  2. 2.School of Computer Science and EngineeringWuhan Institute of TechnologyWuhanChina
  3. 3.SDIC Xinjiang Luobupo Potash Co., Ltd.HamiChina

Personalised recommendations