US 12,761,643 B2Granted

Systems and methods for identifying and predicting industrial machine maintenance events

AssigneeCaterpillar Inc.

InventorsDaniel J. Reaume, Sadia Khalil

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods are disclosed herein for identifying and predicting maintenance events with respect to assets, such as mobile machinery. A computing platform can include control circuit(s) (on-board and/or remote) configured to acquire machine fault data and machine utilization data. A feature set can be generated by transforming the machine fault data (e.g., by fusing the machine fault data with machine utilization data), normalizing the data, generating embeddings, and/or executing imputation algorithms to fill in missing values. A classifier model can be trained using the transformed machine fault data in conjunction with labeled maintenance event data to enable the trained classifier model to learn to recognize utilization-specific fault code patterns. The trained classifier model can generate inferences and predictions regarding machinery maintenance events without the use of maintenance data.

Claims20

Dependent claims

A system for predicting maintenance events for industrial machines, the system comprising:

a set of condition monitoring sensors;
at least one processor;
at least one memory;
and
one or more non-transitory, computer-readable storage media storing instructions, which, when executed by the at least one processor, cause the system to:
generate a training dataset for a machine learning classifier model by:
acquiring, using the set of condition monitoring sensors, a first fault code dataset relating to operation of an industrial machine,
wherein the industrial machine includes one or more of earth-moving machinery, construction machinery, or power-generation machinery;
applying a transformation to the first fault code dataset to fuse fault code data with a first utilization dataset for the industrial machine,
wherein the transformation includes two or more of (i) a fault-to-utilization ratio, (ii) a fault-to-output ratio, or (iii) a ratio change metric,
wherein the training dataset comprises the transformed fault code data;
train the machine learning classifier model using the training dataset and a maintenance dataset labeled with maintenance event data;
acquire a second fault code dataset and a second utilization dataset;
execute the machine learning classifier model on the second fault code dataset and second utilization dataset to generate a maintenance event prediction set;
and
using the maintenance event prediction set, generate a maintenance plan.

13 more elements

One or more non-transitory, computer-readable media having instructions encoded thereon, which, when executed by at least one data processor of a computing system, cause the computing system to perform operations comprising:

generate a training dataset for a machine learning classifier model by:
acquiring, using a set of condition monitoring sensors, a first fault code dataset relating to operation of an industrial machine,
wherein the industrial machine includes one or more of earth-moving machinery, construction machinery, or power-generation machinery;
applying a transformation to the first fault code dataset to fuse fault code data with a first utilization dataset for the industrial machine, wherein the transformation includes two or more of (i) a fault-to-utilization ratio, (ii) a fault-to-output ratio, or (iii) a ratio change metric,
wherein the training dataset comprises the transformed fault code data;
train the machine learning classifier model using the training dataset and a maintenance dataset labeled with maintenance event data;
acquire a second fault code dataset and a second utilization dataset;
execute the machine learning classifier model on the second fault code dataset and second utilization dataset to generate a maintenance event prediction set;
and
using the maintenance event prediction set, generate a maintenance plan.

7 more elements

A computer-implemented method comprising:

generating a training dataset for a machine learning classifier model by:
acquiring, using a set of condition monitoring sensors, a first fault code dataset relating to operation of an industrial machine,
wherein the industrial machine includes one or more of earth-moving machinery, construction machinery, or power-generation machinery;
applying a transformation to the first fault code dataset to fuse fault code data with a first utilization dataset for the industrial machine, wherein the transformation includes two or more of (i) a fault-to-utilization ratio, (ii) a fault-to-output ratio, or (iii) a ratio change metric,
wherein the training dataset comprises the transformed fault code data;
training the machine learning classifier model using the training dataset and maintenance dataset labeled with maintenance event data;
acquiring a second fault code dataset and a second utilization dataset;
executing the machine learning classifier model on the second fault code dataset and second utilization dataset to generate a maintenance event prediction set;
and
using the maintenance event prediction set, generating a maintenance plan.

7 more elements

Dependencies are inferred from the claim text.