Event driven control schemas for artificial lift

Patent metadata

Field	Value
Publication number	US-12221867-B2
Application number	US-202017432363-A
Country	US
Kind code	B2
Filing date	Feb 20, 2020
Priority date	Feb 21, 2019
Publication date	Feb 11, 2025
Grant date	Feb 11, 2025

Abstract

A system for oil and gas production includes a drilling, pipeline, or power quality device, a sensor, and a storage device. The storage device has instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform operations. The operations include receiving first data from the sensor. The operations include determining whether the device is experiencing an event condition based on the first data The operations include generating one or more potential corrective actions responsive to a determination that the device is experiencing the event condition. The operations include calculating a reward value for each of the one or more potential corrective actions. The operations include identifying a first corrective action of the one or more potential corrective actions with a largest or most positive reward value The operations include controlling the device according to the first corrective action.

First claim

What is claimed is:

1. A system for oil and gas production, the system comprising:
a drilling, pipeline, or power quality equipment;
a sensor associated with the drilling, pipeline, or power quality equipment; and
a storage device having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving first data from the sensor, the first data indicating a parameter of the drilling, pipeline, or power quality equipment;
determining whether the drilling, pipeline, or power quality equipment is experiencing an event condition based on the first data;
generating one or more potential corrective actions responsive to a determination that the drilling, pipeline, or power quality equipment is experiencing the event condition, the one or more potential corrective actions each comprising controlling operation of a fluid transfer device, resulting in a change to the parameter of the drilling, pipeline, or power quality equipment, and generating the one or more potential corrective actions comprises:
determining a severity associated with the event condition,
inputting the severity of the event condition and a type of the event condition to a model of the system, and
selecting, from an output of the model, the one or more potential corrective actions, wherein the output of the model is a Gaussian distribution of potential actions;

calculating a reward value for each of the one or more potential corrective actions;
identifying a first corrective action of the one or more potential corrective actions with a largest or a most positive reward value;
controlling the fluid transfer device to implement to the first corrective action comprising sending a first control signal to the fluid transfer device, wherein the first corrective action comprises at least one of increasing a drive frequency of the fluid transfer device, adjusting a choke position of the fluid transfer device, or reversing a rotational direction of an impeller of the fluid transfer device;
determining whether the drilling, pipeline, or power quality equipment is experiencing a subsequent event condition; and
controlling the fluid transfer device to implement a subsequent corrective action comprising sending a second control signal to the fluid transfer device, the subsequent corrective action based on at least a set of subsequent data received from the sensor and the first corrective action.

2. The system of claim 1 , wherein the operations further comprise receiving second data from the sensor associated with the drilling, pipeline, or power quality equipment, the second data received after controlling the fluid transfer device based on the first corrective action, wherein the reward value for the first corrective action is based in part on the second data and wherein the drilling, pipeline, or power quality equipment comprises the fluid transfer device.

3. The system of claim 2 , wherein the operations further comprise:
determining whether the drilling, pipeline, or power quality equipment is still experiencing the event condition based on the second data;
generating a second corrective action based on a determination that the drilling, pipeline, or power quality equipment is still experiencing the event condition; and
controlling the fluid transfer device based on the second corrective action.

4. The system of claim 1 , wherein the operations further comprise training the model according to a predetermined policy.

5. A system for oil and gas production, the system comprising:
a drilling, pipeline, or power quality equipment;
a sensor associated with the drilling, pipeline, or power quality equipment; and
a storage device having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving first data from the sensor, the first data indicating a parameter of the drilling, pipeline, or power quality equipment;
determining whether the drilling, pipeline, or power quality equipment is experiencing an event condition based on the first data, wherein determining whether the drilling, pipeline, or power quality equipment is experiencing the event condition comprises:
determining a likelihood for one or more event conditions associated with the drilling, pipeline, or power quality equipment; and
comparing the likelihood for the one or more event conditions, wherein a condition of the one or more event conditions with a highest likelihood is identified as the event condition;

generating one or more potential corrective actions responsive to a determination that the device is experiencing the event condition, the one or more potential corrective actions each comprising controlling operation of a fluid transfer device, resulting in a change to the parameter of the drilling, pipeline, or power quality device equipment;
calculating a reward value for each of the one or more potential corrective actions;
identifying a first corrective action of the one or more potential corrective actions with a largest or a most positive reward value;
controlling the fluid transfer device to implement the first corrective action comprising sending a first control signal to the fluid transfer device, wherein the first corrective action comprises at least one of increasing a drive frequency of the fluid transfer device, adjusting a choke position of the fluid transfer device, or reversing a rotational direction of an impeller of the fluid transfer device;
determine whether the drilling, pipeline, or power quality equipment is experiencing a subsequent event condition; and

controlling the fluid transfer device to implement a subsequent corrective action, comprising sending a second control signal to the fluid transfer device, the subsequent corrective action based on at least a set of subsequent data received from the sensor and the first corrective action.

6. The system of claim 1 , wherein calculating the reward value for each of the one or more potential corrective actions further comprises determining a past reward value and a future reward value for each of the one or more potential corrective actions, wherein the reward value for each of the one or more potential corrective actions is based in part on the past reward value and the future reward value.

7. The system of claim 1 , wherein the one or more potential corrective actions are generated according to a constraint that limits the change to the parameter of the drilling, pipeline, or power quality equipment.

8. The system of claim 1 , wherein the operations further comprise determining an impact of each of the one or more potential corrective actions on a likelihood for each of one or more other event conditions, wherein the reward value for each of the one or more potential corrective actions is based in part on the impact.

9. A method for automatic event detection and correction for oil and gas production equipment, the method comprising:
receiving first data from a sensor associated with the oil and gas production equipment, the first data indicating a parameter of the oil and gas production equipment;
determining whether the oil and gas production equipment is experiencing an event condition based on the first data;
generating one or more potential corrective actions responsive to a determination that the oil and gas production equipment is experiencing the event condition, the one or more potential corrective actions each comprising controlling operation of a fluid transfer device, resulting in a change to the parameter of the oil and gas production equipment, wherein generating the one or more potential corrective actions further comprises:

Assignees

• Sensia Llc
• Schlumberger Technology Corp

Inventors

• Chong Jonathan Wun Shiung
• Chambliss T Everett
• Hedlund Magnus
• Rossi David
• Nguyen Nam
• Yuan Pengyu
• Hoefel Albert

Classifications

• E21B2200/20

  Computer models or simulations, e.g. for reservoirs under production, drill bits · CPC title

• E21B47/008 Primary

  Monitoring of down-hole pump systems, e.g. for the detection of "pumped-off" conditions · CPC title

• G06Q50/02

  Agriculture; Fishing; Forestry; Mining · CPC title

• G06Q10/0637

  Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals · CPC title

• E21B43/121 Primary

  Lifting well fluids (monitoring of down-hole pump systems E21B47/008) · CPC title