Predicting the Efficiency of the Oil Removal From Surfactant and Polymer Produced Water by Using Liquid–Liquid Hydrocyclone: Comparison of Prediction Abilities Between Response Surface Methodology and Adaptive Neuro-Fuzzy Inference System

The present study developed, evaluated and compared the prediction and simulating efficiency of both, the response surface methodology (RSM) and adaptive neuro-fuzzy inference system (ANFIS) approaches for oil removal using a liquid-liquid hydrocyclone (LLHC) from surfactant and polymer (SP) produced water. Six parameters were involved in the process: the surfactant concentration, polymer concentration, salinity, initial oil concentration, feed flowrate and split ratio. For RSM, D-optimal design was used, while the ANFIS model was developed in term of this process with the Gaussian membership function. All models were compared statistically based on the training and testing data set by the coefficient of determination (R2), root-mean-square error (RMSE), average absolute percentage error (AAPE), standard deviation (STD), minimum error, and maximum error. The R2 for RSM and the ANFIS model for the testing set were of 0.972 and 0.999, respectively. Both models made good predictions. Trend analysis has been done to confirm the applicability of the models. From the results, it shows that the ANFIS model was more precise compared to the RSM model, which proves that the ANFIS is a powerful tool for modelling and optimizing the efficiency of the oil removal from the LLHC in the presence of SP.

[2]  Kok Wai Wong,et al.  Hybrid fuzzy modelling using memetic algorithm for hydrocyclone control , 2004, Proceedings of 2004 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.04EX826).

[3]  I. C. Smyth,et al.  Development and performance of oil-water hydrocyclone separators: a review , 1998 .

[4]  G. Sams,et al.  Challenges in Processing Produced Emulsion from Chemical Enhanced Oil Recovery - Polymer Flood Using Polyacrylamide , 2011 .

[5]  Zhenyu Yang,et al.  Evaluation of OiW measurement technologies for deoiling hydrocyclone efficiency estimation and control , 2016, OCEANS 2016 - Shanghai.

[6]  Henglin Yang,et al.  Effect of viscosity and interfacial tension of surfactant–polymer flooding on oil recovery in high-temperature and high-salinity reservoirs , 2014, Journal of Petroleum Exploration and Production Technology.

[7]  Abu Azam Md. Yassin.  Legislation On Oil Pollution Prevention And Control During Petroleum Production , 1988 .

[8]  G. Chen,et al.  Produced water treatment technologies , 2014 .

[9]  Shoubo Wang,et al.  Oil-Water Separation in Liquid-Liquid Hydrocyclones (LLHC) -Experiment and Modeling , 2001 .

[10]  Gang Yu,et al.  Effects of alkaline/surfactant/polymer on stability of oil droplets in produced water from ASP flooding , 2002 .

[11]  Luiz Gustavo Martins Vieira,et al.  Optimization of Design and Performance of Solid‐Liquid Separators: A Thickener Hydrocyclone , 2015 .

[12]  Ming Chen,et al.  Separation Performance of a Novel Liquid–Liquid Dynamic Hydrocyclone , 2018 .

[13]  R. Steiner,et al.  D-optimal experimental designs for Freundlich and Langmuir adsorption isotherms , 2009 .

[14]  G. van Schoor,et al.  Hydrocyclone cut-size estimation using artificial neural networks , 2016 .

[15]  N. Meldrum,et al.  Hydrocyclones: a solution to produced-water treatment , 1988 .

[16]  Tamás D. Gedeon,et al.  Fuzzy rule interpolation for multidimensional input spaces in determining d50c of hydrocyclones , 2003, IEEE Trans. Instrum. Meas..

[17]  Halit Eren,et al.  Developing a generalised neural-fuzzy hydrocyclone model for particle separation , 1998, IMTC/98 Conference Proceedings. IEEE Instrumentation and Measurement Technology Conference. Where Instrumentation is Going (Cat. No.98CH36222).

[18]  Shishi Pang,et al.  Properties of Emulsions Formed In Situ In a Heavy-Oil Reservoir during Water Flooding: Effects of Salinity and pH , 2018, Journal of Surfactants and Detergents.

[19]  Min Yang,et al.  The effects of oil displacement agents on the stability of water produced from ASP (alkaline/surfactant/polymer) flooding , 2011 .

[20]  B. E. Bowers,et al.  Development of a Downhole Oil/Water Separation and Reinjection System for Offshore Application , 2000 .

[21]  Torleiv Bilstad,et al.  Operational Control of Hydrocyclones During Variable Produced Water Flow Rates—Frøy Case Study , 2007 .

[22]  S. K. Nicol,et al.  Concentration of oil-in-water emulsion using the air-sparged hydrocyclone , 1993 .

[23]  Lyes Khezzar,et al.  Hydrocyclones for De-oiling Applications—A Review , 2010 .

[24]  Mohsen Karimi,et al.  Prediction of hydrocyclone performance using artificial neural networks , 2010 .

[25]  R. Yunus,et al.  Stability Investigation of Water-in-Crude Oil Emulsion , 2006 .

[26]  Gursharan Singh,et al.  Artificial Neuro-Fuzzy Inference System (ANFIS) based validation of laccase production using RSM model , 2018 .

[27]  Adem Yavuz Sönmez,et al.  An Adaptive Neuro-Fuzzy Inference System (ANFIS) to Predict of Cadmium (Cd) Concentrations in the Filyos River, Turkey , 2018 .

[28]  Ahmad Hanif Asyhar,et al.  Forecasts marine weather on Java sea using hybrid methods: TS-ANFIS , 2017, 2017 4th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI).

[29]  E. Khor.  Improvements of oil-in-water analysis for produced water using membrane filtration , 2011 .

[31]  M. Sillanpää,et al.  Neuro-fuzzy modeling to adsorptive performance of magnetic chitosan nanocomposite , 2017, Journal of Nanostructure in Chemistry.

[32]  Chuen-Tsai Sun,et al.  Neuro-fuzzy modeling and control , 1995, Proc. IEEE.

[33]  Other.  Directive 2000/60/EC of the European Parliament and of The Council of 23 October 2000 establishing a Framework for Community Action in the Field of Water Policy (Water Framework Directive) , 2000 .

[34]  Produced-Water-Treatment Systems : Comparison of North Sea and Deepwater Gulf of Mexico , 2015 .

[35]  H. Al-Kayiem,et al.  Evaluation of Alkali/Surfactant/Polymer Flooding on Separation and Stabilization of Water/Oil Emulsion by Statistical Modeling , 2017 .

[36]  Becky Turner,et al.  New Water-Treatment Technologies Tackle Offshore Produced-Water Challenges in EOR , 2013 .

[37]  Kemal Maulana Alhasa,et al.  Modeling of Tropospheric Delays Using ANFIS , 2015 .

[38]  Chris Lacor,et al.  Modeling and Pareto optimization of gas cyclone separator performance using RBF type artificial neural networks and genetic algorithms , 2012 .

[39]  Zhenyu Yang,et al.  Experimental modeling of a deoiling hydrocyclone system , 2015, 2015 20th International Conference on Methods and Models in Automation and Robotics (MMAR).

[40]  Tormod Drengstig,et al.  Performance of a deoiling hydrocyclone during variable flow rates , 2007 .

[41]  Lotfi A. Zadeh,et al.  Outline of a New Approach to the Analysis of Complex Systems and Decision Processes , 1973, IEEE Trans. Syst. Man Cybern..

[42]  A. B. Sinker,et al.  Enhanced Deoiling Hydrocyclone Performance without Resorting to Chemicals , 1999 .

[43]  Jyh-Shing Roger Jang,et al.  ANFIS: adaptive-network-based fuzzy inference system , 1993, IEEE Trans. Syst. Man Cybern..