US 11,719,683 B2
Automated real-time water cut testing and multiphase flowmeter calibration advisory
Ali M. Al Shahri, Dhahran (SA); Anas A. Al Shuaibi, Dammam (SA); Mohammed Sami Kanfar, Dammam (SA); Kalid Saad Dosary, Al-Khobar (SA); and Abdulaziz A. Alsaleh, Dammam (SA)
Assigned to Saudi Arabian Oil Company, Dhahran (SA)
Filed by Saudi Arabian Oil Company, Dhahran (SA)
Filed on Mar. 29, 2021, as Appl. No. 17/215,591.
Claims priority of provisional application 63/002,469, filed on Mar. 31, 2020.
Prior Publication US 2021/0302405 A1, Sep. 30, 2021
Int. Cl. G01N 33/28 (2006.01); E21B 47/07 (2012.01); E21B 47/06 (2012.01)
CPC G01N 33/2847 (2013.01) [E21B 47/06 (2013.01); E21B 47/07 (2020.05); G01N 33/2823 (2013.01); E21B 2200/20 (2020.05); E21B 2200/22 (2020.05)] 18 Claims
OG exemplary drawing
 
1. An intelligent water cut estimation system comprising at least two pressure sensors, a neural network model, and a data processor, wherein:
the at least two pressure sensors are configured to generate pressure data respectively associated with two points of a well bore;
the neural network model comprises one or more parameters indicative of water cut associated with a well bore; and
the data processor is communicatively coupled to the at least two pressure sensors and the neural network model and is operable to:
receive pressure data from the at least two pressure sensors respectively indicative of the pressure at each of the two points of the well bore,
determine a pressure drop between the two points based on the received pressure data from the at least two pressure sensors,
generate an input water cut estimate,
estimate a dynamic pressure loss based on the pressure drop, the one or more parameters of the neural network model, and the input water cut estimate to initiate an iterative process,
estimate a potential energy loss based on the dynamic pressure loss and the pressure drop,
inverse model a water cut estimate based on the potential energy loss,
compare the water cut estimate to the input water cut estimate to generate a water cut Δ,
utilize the water cut estimate as the input water cut estimate for the iterative process when the water cut Δ exceeds a threshold, and
continue the iterative process until the water cut Δ is below the threshold.