US 12,546,207 B2Granted

Method, analysis device and pumping system for detecting an oil leak in a beam pump

AssigneeSiemens Energy Global GmbH & Co. KG

InventorsJürgen Hatzl, Severin Küberl, Martin Mayer, Stefan Wakolbinger

Life of this patent

Estimating the term…

Loading patent…

Abstract

A computer-implemented method for detecting an oil leak in a beam pump, wherein, during pumping movement of the rod linkage through the wellhead, a stroke region is formed, the method including a) capturing images of an image sequence containing at least the stroke region of the rod linkage, b) detecting soiling in the stroke region of the rod linkage in comparison with a known clean state for an image in question of the image sequence, c) ascertaining the position of the detected soiling for an image in question of the image sequence by the associated position of the rod linkage, d) determining the change in the soiling at the ascertained position over at least one pumping cycle from the images of the image sequence, e) checking if the change in the soiling exceeds a predefined limit value, and if applicable, outputting a warning of an oil leak.

Claims19

Dependent claims

A computer-implemented method for identifying an oil escape in a beam pump, wherein the pump has a pump head which is connected via a linkage to a kinematic converter and the kinematic converter is driven during operation by a motor, and the linkage is inserted into a wellhead located in the ground, characterized in that during a pumping movement of the linkage through the wellhead, a stroke region is formed which is captured by a camera via an image sequence, the method comprising:

a) capturing images of the image sequence with at least the stroke region of the linkage,
b) identifying fouling in the stroke region of the linkage in comparison with a known clean state for a respective image of the image sequence,
c) ascertaining the location of the identified fouling for a respective image of the image sequence with the aid of the respective position of the linkage,
d) determining a variation of the fouling at the ascertained location over at least one pump cycle from the images of the image sequence,
e) checking whether the variation of the fouling exceeds a predetermined limit value and, if appropriate, emitting a warning of an oil escape,
wherein a reference marking is applied on the linkage and the location of the identified fouling in relation to the location of the reference marking is ascertained by the reference marking being captured at least partially by the camera via images of the image sequence;
and
wherein timestamps for the images of the image sequence are respectively captured at an acquisition instant, and the location of the identified fouling in relation to the location of the reference marking is ascertained for the images of the image sequence by means of the timestamps.

5 more elements

a) capturing images of the image sequence with at least the stroke region of the linkage,
b) identifying fouling in the stroke region of the linkage in comparison with a known clean state for a respective image of the image sequence,
c) ascertaining the location of the identified fouling for a respective image of the image sequence with the aid of the respective position of the linkage,
d) determining a variation of the fouling at the ascertained location over at least one pump cycle from the images of the image sequence,
e) checking whether the variation of the fouling exceeds a predetermined limit value and, if appropriate, emitting a warning of an oil escape,
wherein the motor is an electric motor and the location of the identified fouling in relation to a position of the linkage is ascertained by a power consumption of the motor being captured with a sensor means during a pumping movement, from which the position of the linkage is ascertained.

3 more elements

a) capturing images of the image sequence with at least the stroke region of the linkage,
b) identifying fouling in the stroke region of the linkage in comparison with a known clean state for a respective image of the image sequence,
c) ascertaining the location of the identified fouling for a respective image of the image sequence with the aid of the respective position of the linkage,
d) determining a variation of the fouling at the ascertained location over at least one pump cycle from the images of the image sequence,
e) checking whether the variation of the fouling exceeds a predetermined limit value and, if appropriate, emitting a warning of an oil escape,
wherein a model of a shape and/or a texture and a respective time variation thereof is generated, trained and applied on the basis of machine learning when determining the variation of the fouling.

3 more elements

a) capturing images of the image sequence with at least the stroke region of the linkage,
b) identifying fouling in the stroke region of the linkage in comparison with a known clean state for a respective image of the image sequence,
c) ascertaining the location of the identified fouling for a respective image of the image sequence with the aid of the respective position of the linkage,
d) determining a variation of the fouling at the ascertained location over at least one pump cycle from the images of the image sequence,
e) checking whether the variation of the fouling exceeds a predetermined limit value and, if appropriate, emitting a warning of an oil escape,
wherein the location of the identified fouling in relation to a position of the linkage is ascertained by a movement at the linkage or at the kinematic converter being captured by at least one sensor, from which the position of the linkage is ascertained.

3 more elements

Dependencies are inferred from the claim text.

Facts

Application: 18/032,181Filed 19 Oct 2021
Priority: EP 2020333822 Oct 2020
Published: US 2023/0383642 A130 Nov 2023
Granted: 10 Feb 2026
Term ends: Not estimatedThe source records don’t support an estimate.

Classification6 codes

E21B 47/009
Monitoring of down-hole pump systems, e.g. for the detection of "pumped-off" conditions: monitoring of walking-beam pump systems
G06V 10/25
Image preprocessing: determination of region of interest [ROI] or a volume of interest [VOI]
G06V 20/52
Context or environment of the image: surveillance or monitoring of activities, e.g. for recognising suspicious objects
E21B 2200/22
Special features related to earth drilling for obtaining oil, gas or water: fuzzy logic, artificial intelligence, neural networks or the like