The formation of inorganic scale, particularly calcium carbonate (CaCO3), is a persistent and one of the most serious and costly problems in the oil and gas industries. This event may cause partial to complete plugging, block valves, tubing and flowlines, and then reduce the production rates. This article proposes the use of support vector regression to build a nonlinear mapping between a set of variables (surface cladding, material, temperature, pressure, brine composition, and fluid velocity) and the scale build-up. The support vector regression is fed with data gathered from laboratory tests carried out on coupons that simulate realistic downhole conditions encountered in oil well bores from the pre-salt fields in Brazil. The proposed failure prediction framework is comprehensive as it entails the stages of hyperparameter tuning, variable selection, and uncertainty analysis, which are addressed by a combination of particle swarm optimization and bootstrap with support vector regression. The obtained results suggest that the bootstrapped particle swarm optimization + support vector regression is a valuable tool that may be used to support condition-based maintenance-related decisions.
DyerSJGrahamGM. The effect of temperature and pressure on oilfield scale formation. J Petrol Sci Eng2002; 35: 95–107.
2.
ChenTNevilleAYuanM. Calcium carbonate scale formation—assessing the initial stages of precipitation and deposition. J Petrol Sci Eng2005; 46: 185–194.
3.
ZhangYShawHFarquharR. The kinetics of carbonate scaling—application for the prediction of downhole carbonate scaling. J Petrol Sci Eng2001; 29: 85–95.
4.
HamidSDe JesusOJacintoCMC. A practical method of predicting chemical scale formation in well completions. In: SPE SAS annual technical symposium & exhibition, Khobar, Saudi Arabia: Society of Petroleum Engineers, 19–22 May 2013.
5.
SchölkopfBSmolaAJ. Learning with kernels: support vector machines, regularization, optimization and beyond. Cambridge, MA: The MIT Press, 2002.
6.
BoydSVandenbergheL. Convex optimization. Cambridge: Cambridge University Press, 2004.
7.
MommaMBennettKP. A pattern search method for model selection of support vector regression. In: Proceedings of the 2nd SIAM international conference on data mining, Arlington, VA: SIAM (Society of Industrial and Applied Mathematics), 11–13 April 2002, pp.261–274.
8.
ChapelleOVapnikVBousquetO. Choosing multiple parameters for support vector machines. Mach Learn2002; 46: 131–159.
9.
ItoKNakanoR. Optimizing support vector regression hyperparameters based on cross-validation. In: Proceedings of the international joint conference on neural networks, Portland, OR, 20–24 July 2003, pp.2077–2082.
10.
ChangM-WLinC-J. Leave-one-out bounds for support vector regression model selection. Neural Comput2005; 17: 1188–1222.
11.
PaiP-F. System reliability forecasting by support vector machines with genetic algorithms. Math Comput Model2006; 43: 262–274.
12.
ChenK-Y. Forecasting systems reliability based on support vector regression with genetic algorithms. Reliab Eng Syst Safe2007; 92: 423–432.
13.
LinS-WYingK-CChenS-C. Particle swarm optimization for parameter determination and feature selection of support vector machines. Expert Syst Appl2008; 35: 1817–1824.
14.
FeiS-WWangM-JMiaoY-B. Particle swarm optimization-based support vector machine for forecasting dissolved gases content in power transformer oil. Energ Convers Manage2009; 50: 1604–1609.
15.
LinsIDMouraMCZioE. A particle swarm-optimized support vector machine for reliability prediction. Qual Reliab Eng Int2012; 28: 141–158.
16.
BaraldiPDi MaioFZioE. Sensitivity analysis of scale deposition on equipment of oil wells plants. Chem Eng Trans2012; 26: 327–332.
17.
GuyonIElisseeffA. An introduction to variable and feature selection. J Mach Learn Res2003; 3: 1157–1182.
18.
LinsIDMouraMCDroguettEL. Reliability prediction of oil wells by support vector machine with particle swarm optimization for variable selection and hyperparameter tuning. In: European Safety & Reliability Association Conference (ESREL), Troyes, 18–22 September 2011. Advances in safety, reliability and risk management - Bérenguer, C.; GrallA.Guedes SoaresC. (eds.). Londres: CRC Press / Balkema - Taylor & Francis Group, 2011. pp. 1488–1496.
19.
LinsIDMouraMCDroguettEA. Failure prediction of oil wells by support vector regression with variable selection, hyperparameter tuning and uncertainty analysis. Chem Eng Trans2013; 33: 817–822.
20.
BrabanterKDBrabanterJDSuykensJAK. Approximate confidence and prediction intervals for least squares support vector regression. IEEE T Neural Networ2011; 22: 110–120.
21.
EfronB. Bootstrap methods: another look at the jackknife. Ann Stat1979; 7: 1–26.
22.
EfronBTibshiraniRJ. An introduction to the bootstrap. New York: Chapman & Hall/CRC, 1993.
23.
KecmanV. Support vector machines: an introduction. In: WangL (ed.) Support vector machines: theory and applications. Berlin, Heidelberg: Springer-Verlag, 2005, pp.1–47.
24.
HsuC-WChangC-CLinC-J. A practical guide to support vector classification. Report, Department of Computer Science, National Taiwan University, Taiwan, 2003, http://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf (accessed 17 March 2014).
25.
LinH-TLinC-J. A study on sigmoid kernels for SVM and the training of non-PSD kernels by SMO-type methods. Report, Department of Computer Science, National Taiwan University, Taiwan, 2003, http://home.caltech.edu/~htlin/publication/doc/tanh.pdf (accessed 17 March 2014).
26.
BrattonDKennedyJ. Defining a standard for particle swarm optimization. In: IEEE Swarm Intelligence Symposium, Honolulu, HI, 1–5 April 2007, pp.120–127.
27.
DavisonACHinkleyDV. Bootstrap methods and their applications. New York: Cambridge University Press, 1997.
28.
LinsIDMouraMCDroguettEL. Uncertainty assessment through bootstrapped support vector regression. In: Proceedings of the 11th international probabilistic and safety assessment and management conference (PSAM 11) and European Safety & Reliability Conference (ESREL), Helsinki, 25–29 June 2012, Red Hook, NY, Estados Unidos: Curran Associates, Inc., 2012. v. 2. pp. 908–917.
29.
LiuR. Bootstrap procedures under some non-I.I.D. models. Ann Stat1988; 16: 1696–1708.
30.
DavidsonJMonticiniAPeelD. Implementing the wild bootstrap using a two-point distribution. Econ Lett2007; 96: 309–315.
