Back to results
US 11,144,378 B2Granted
Computer system and method for recommending an operating mode of an asset
AssigneeUptake Technologies, Inc.
InventorsStephanie Shapiro, Brian Silva
Life of this patent
Estimating the term…
Abstract
Disclosed herein are systems, devices, and methods related to assets and asset operating conditions. In particular, examples involve determining health metrics that estimate the operating health of an asset or a part thereof, determining recommended operating modes for assets, analyzing health metrics to determine variables that are associated with high health metrics, and modifying the handling of operating conditions that normally result in triggering of abnormal-condition indicators, among other examples.
Claims20
A computing system comprising:
- a network interface configured to interface the computing system with a communication network;
- at least one processor;
- a non-transitory computer-readable medium;
- and
- program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor to cause the computing system to:
- receive sensor data for a sensor-equipped asset, wherein the sensor data indicates operating conditions for the sensor-equipped asset;
- input the sensor data for the sensor-equipped asset into a plurality of individual failure models for a group of failure types, wherein each individual failure model functions to (1) receive the sensor data for the sensor-equipped asset as input and (2) output a respective value indicating a likelihood of a respective failure type occurring at the sensor-equipped asset within a future period of time, wherein the individual failure model for each respective failure type in the group comprises a machine learning model that was defined by applying a machine learning technique to sensor data corresponding to instances of the respective failure type that occurred at assets in the past;
- based on the respective value output by each of the plurality of individual failure models for the group of failure types, determine that two or more different failure types are predicted to occur at the sensor-equipped asset;
- identify a respective categorization of each of the two or more different failure types, wherein identifying the respective categorization of each of the two or more different failure types results in an identification of at least two different categorization levels along a preestablished scale of categorization levels;
- based on the respective categorization of each of the two or more different failure types, determine one particular recommended operating mode of the sensor-equipped asset that comprises a recommendation of a particular capacity in which the sensor-equipped asset should be used during the future period of time, wherein determining the one particular recommended operating mode comprises:
- using preestablished criteria to select, from the at least two different categorization levels along the preestablished scale of categorization levels, one single categorization level that serves as a representative categorization for the two or more different failure types, wherein the one single categorization level corresponds to one single recommended operating mode along a preestablished scale of recommended operating modes;
- and
- identifying the one single recommended operating mode that corresponds to the one single categorization level as the one particular recommended operating mode of the sensor-equipped asset;
- and
- cause a computing device to display a visual representation of the one particular recommended operating mode of the sensor-equipped asset.
12 more elements
A non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium is provisioned with software that is executable to cause a computing system to perform functions including:
- receiving sensor data for a sensor-equipped asset, wherein the sensor data indicates operating conditions for the sensor-equipped asset;
- inputting the sensor data for the sensor-equipped asset into a plurality of individual failure models for a group of failure types, wherein each individual failure model functions to (1) receive the sensor data for the sensor-equipped asset as input and (2) output a respective value indicating a likelihood of a respective failure type occurring at the sensor-equipped asset within a future period of time, wherein the individual failure model for each respective failure type in the group comprises a machine learning model that was defined by applying a machine learning technique to sensor data corresponding to instances of the respective failure type that occurred at assets in the past;
- based on the respective value output by each of the plurality of individual failure models for the group of failure types, determining that two or more different failure types are predicted to occur at the sensor-equipped asset;
- identifying a respective categorization of each of the two or more different failure types, wherein identifying the respective categorization of each of the two or more different failure types results in an identification of at least two different categorization levels along a preestablished scale of categorization levels;
- based on the respective categorization of each of the two or more different failure types, determining one particular recommended operating mode of the sensor-equipped asset that comprises a recommendation of a particular capacity in which the sensor-equipped asset should be used during the future period of time, wherein determining the one particular recommended operating mode comprises:
- using preestablished criteria to select, from the at least two different categorization levels along the preestablished scale of categorization levels, one single categorization level that serves as a representative categorization for the two or more different failure types, wherein the one single categorization level corresponds to one single recommended operating mode along a preestablished scale of recommended operating modes;
- and
- identifying the one single recommended operating mode that corresponds to the one single categorization level as the one particular recommended operating mode of the sensor-equipped asset;
- and
- causing a computing device to display a visual representation of the one particular recommended operating mode of the sensor-equipped asset.
7 more elements
A computer-implemented method comprising:
- receiving sensor data for a sensor-equipped asset, wherein the sensor data indicates operating conditions for the sensor-equipped asset;
- inputting the sensor data for the sensor-equipped asset into a plurality of individual failure models for a group of failure types, wherein each individual failure model functions to (1) receive the sensor data for the sensor-equipped asset as input and (2) output a respective value indicating a likelihood of a respective failure type occurring at the sensor-equipped asset within a future period of time, wherein the individual failure model for each respective failure type in the group comprises a machine learning model that was defined by applying a machine learning technique to sensor data corresponding to instances of the respective failure type that occurred at assets in the past;
- based on the respective value output by each of the plurality of individual failure models for the group of failure types, determining that two or more different failure types are predicted to occur at the sensor-equipped asset;
- identifying a respective categorization of each of the two or more different failure types, wherein identifying the respective categorization of each of the two or more different failure types results in an identification of at least two different categorization levels along a preestablished scale of categorization levels;
- based on the respective categorization of each of the two or more different failure types, determining one particular recommended operating mode of the sensor-equipped asset that comprises a recommendation of a particular capacity in which the sensor-equipped asset should be used during the future period of time, wherein determining the one particular recommended operating mode comprises:
- using preestablished criteria to select, from the at least two different categorization levels along the preestablished scale of categorization levels, one single categorization level that serves as a representative categorization for the two or more different failure types, wherein the one single categorization level corresponds to one single recommended operating mode along a preestablished scale of recommended operating modes;
- and
- identifying the one single recommended operating mode that corresponds to the one single categorization level as the one particular recommended operating mode of the sensor-equipped asset;
- and
- causing a computing device to display a visual representation of the one particular recommended operating mode of the sensor-equipped asset.
7 more elements
Dependencies are inferred from the claim text.