Back to results
US 11,962,957 B1Granted
Systems and methods for wellsite control
AssigneeSchlumberger Technology Corporation
InventorsJulien Converset, Emmanuel Coste, Graeme Paterson
Life of this patent
Estimating the term…
Abstract
A method for wellsite control includes, at a computing device, receiving sensor information from one or more sensors; determining at least one notification option based at least partially on the sensor information; selecting at least one notification option based at least partially on the sensor information; and sending a notification to a notification destination.
Claims17
A method for wellsite control, the method comprising:
- at a computing device:
- receiving primary sensor information from one or more first sensors;
- inputting the primary sensor information into a machine learning (ML) model, wherein the ML model is configured to receive the primary sensor information and, based on prior training datasets, determine one or more notification options based at least partially on the primary sensor information;
- determining, via the ML model, the one or more notification options based at least partially on the primary sensor information;
- selecting at least one notification option from the one or more notification options based at least partially on the primary sensor information;
- and
- sending a notification to a notification destination;
- receiving a confirmation of the notification from the notification destination, wherein the confirmation of the notification includes an indication of an action taken in conjunction with the notification;
- recording the confirmation of the notification in the ML model;
- training the ML model based on the confirmation of the notification;
- receiving secondary sensor information from one or more second sensors, wherein the secondary sensor information is based at least partially on the primary sensor information;
- inputting the secondary sensor information into the ML model, wherein the ML model is configured to receive the secondary sensor information and the primary sensor information and, based on the prior training datasets, determine the one or more notification options in response based at least partially on the secondary sensor information;
- and
- determining, via the ML model, the one or more notification options based at least partially on the secondary sensor information.
11 more elements
A method for wellsite control, the method comprising:
- at a computing device:
- receiving primary sensor information from one or more first sensors;
- inputting the primary sensor information into a machine learning (ML) model, wherein the ML model is configured to receive the primary sensor information and, based on prior training datasets, determine one or more notification options based at least partially on the primary sensor information;
- determining, via the ML model, the one or more notification options based at least partially on the primary sensor information;
- providing for display of a graphical interface, wherein the graphical interface comprises a first sub-window showing the primary sensor information and the graphical interface further comprises a second sub-window showing the one or more notification options based at least partially on the primary sensor information;
- receiving, via the graphical interface, input data indicating a suggested notification from the one or more notification options;
- sending the suggested notification to a notification destination;
- receiving a confirmation of the suggested notification from the notification destination, wherein the confirmation of the suggested notification includes an indication of an action taken in conjunction with the suggested notification;
- recording the confirmation of the suggested notification in the ML model;
- and
- training the ML model based on the confirmation of the suggested notification;
- receiving secondary sensor information from one or more second sensors, wherein the secondary sensor information is based at least partially on the primary sensor information;
- inputting the secondary sensor information into the ML model, wherein the ML model is configured to receive the secondary sensor information and the primary sensor information and, based on the prior training datasets, determine the one or more notification options in response based at least partially on the secondary sensor information;
- and
- determining, via the ML model, the one or more notification options based at least partially on the secondary sensor information.
12 more elements
A system for wellsite control, the system comprising:
- at least one sensor located at a wellsite;
- a computing device in data communication with the at least one sensor, the computing device including:
- a processor,
- a hardware storage device having instructions stored thereon, that when executed by the processor, cause the computing device to:
- receive sensor information from the at least one sensor;
- input the sensor information into a machine learning (ML) model, wherein the ML model is configured to receive the sensor information and, based on prior training datasets, determine one or more notification options based at least partially on the sensor information;
- determine, via the ML model, the one or more notification options based at least partially on the sensor information;
- provide for display of a graphical interface, wherein the graphical interface comprises a first sub-window showing the sensor information and the graphical interface further comprises a second sub-window showing the one or more notification options based at least partially on the sensor information;
- receive, via the graphical interface, input data indicating a suggested notification from the one or more notification options;
- send the suggested notification to a notification destination;
- receive a confirmation of the suggested notification from the notification destination, wherein the confirmation of the suggested notification includes an indication of an action taken in conjunction with the suggested notification;
- record the confirmation of the suggested notification in the ML model:
- and
- train the ML model based on the confirmation of the suggested notification;
- receive secondary sensor information from one or more second sensors, wherein the secondary sensor information is based at least partially on the sensor information;
- input the secondary sensor information into the ML model, wherein the ML model is configured to receive the secondary sensor information and the sensor information and, based on the prior training datasets, determine the one or more notification options in response based at least partially on the secondary sensor information;
- and
- determine, via the ML model, the one or more notification options based at least partially on the secondary sensor information.
15 more elements
Dependencies are inferred from the claim text.