Flow Measurement and Instrumentation 76 (2020) 101849 Available online 12 November 2020 0955-5986/© 2020 Elsevier Ltd. All rights reserved. Adaptive neuro-fuzzy algorithm applied to predict and control multi-phase flow rates through wellhead chokes Hamzeh Ghorbani a , David A. Wood b, * , Nima Mohamadian c , Sina Rashidi d , Shadfar Davoodi e , Alireza Soleimanian f , Amirafzal Kiani Shahvand d , Mohammad Mehrad g a Young Researchers and Elite Club, Ahvaz Branch, Islamic Azad University, Ahvaz, Iran b DWA Energy Limited, Lincoln, United Kingdom c Young Researchers Club, Petroleum Department, Azad University, Omidiyeh Branch, Iran d Young Researchers and Elite Club, Omidiyeh Branch, Islamic Azad University, Omidiyeh, Iran e School of Earth Science and Engineering, Tomsk Polytechnic University, Lenin Avenue, Tomsk, Russia f Department of Chemical and Petroleum Engineering, Sharif University of Technology, Azadi Ave., Tehran, Iran g Faculty of Mining, Petroleum and Geophysics Engineering, Shahrood University of Technology, Shahrood, Iran A R T I C L E INFO Keywords: Multi-phase oil Gas Water flow rate Fuzzy machine learning Wellhead choke variables Empirical relationships Takagi-Sugeno fuzzy inference system Fuzzy system control ABSTRACT A Takagi-Sugeno adaptive neuro-fuzzy inference system (TSFIS) model is developed and applied to a dataset of wellhead flow-test data for the Resalat oil field located offshore southern Iran, the objective is to assist in the prediction and control of multi-phase flow rates of oil and gas through the wellhead chokes. For this purpose, 182 test data points (Appendix 1) related to the Resalat field are evaluated. In order to predict production flow rate (Q L ) expressed as stock-tank barrels per day (STB/D), this dataset includes four selected input variables: up- stream pressure (Pwh); wellhead choke sizes (D64); gas to liquid ratio (GLR); and, base solids and water including some water-soluble oil emulsion (BS&W). The test data points evaluated include a wide range of oil flow rate conditions and values for the four input variables recorded. The TSFIS algorithm applied involves five data processing steps: a) pre-processing, b) fuzzification, c) rules base and adaptive neuro-fuzzy inference engine, d) defuzzification, and e) post-processing of the fuzzy model. The developed TSFIS model for the Resalat oil field database predicted oil flow rate to a high degree of accuracy (root mean square error = 247 STB/D, correlation coefficient = 0.9987), which improves substantially on the commonly used empirical algorithms used for such predictions. TSFIS can potentially be applied in wellhead choke fuzzy controllers to stabilize flow in specific wells based on real-time input data records. 1. Introduction Accurate determination and estimation of the flow rates of oil and gas production from wells is critical for several purposes [1]. These include material balance calculations, monitoring and evaluating reservoir performance, flow regimes and identifying the fluids states in the reservoir. Specifically, flow rate information is required to establish changes in pressure (ΔP), well-flow pressures (P wf ) average pressure in the reservoir (Pr), choke-performance relationships (CPR), liquid-holdup and inflow-performance relationships (IPR) and the determination of production tubing performances (TPR). Consequently, reservoir engineers continuously strive to improve the estimation ac- curacy of production flow rates based on available information gained from wellhead measurements. The more accurately the flow rate can be determined the more confidence that can be attached to the values of reservoir properties that depend upon it [2]. Therefore, the accurate determination of production flow rates of all fluid phases flowing through wellheads has become a priority for reservoir engineers [3–5]. The wellhead flowrate is usually controlled by well head chokes. Chokes (Appendix 1) are devices routinely installed on oil and gas wellheads in order to control flow rate to desired levels. They also help safeguard wellhead equipment [6], prevent water and gas conning in the reservoir, create back pressure on * Corresponding author. E-mail addresses: hamzehghorbani68@yahoo.com (H. Ghorbani), dw@dwasolutions.com (D.A. Wood), nima.0691@gmail.com (N. Mohamadian), sinarashidi2014@gmail.com (S. Rashidi), davoodis@hw.tpu.ru (S. Davoodi), alireza.soleimanian94@gmail.com (A. Soleimanian), kiani.petro215@gmail.com (A.K. Shahvand), mmehrad1986@gmail.com (M. Mehrad). Contents lists available at ScienceDirect Flow Measurement and Instrumentation journal homepage: http://www.elsevier.com/locate/flowmeasinst https://doi.org/10.1016/j.flowmeasinst.2020.101849 Received 3 April 2020; Received in revised form 23 September 2020; Accepted 23 October 2020