Anomaly detection for refrigeration systems
Methods and systems are described for anomaly detection in refrigeration systems. A process for providing anomaly detection for refrigeration systems includes receiving telemetry data of one or more refrigeration systems, including measured temperature values and setpoint temperature values; processing the telemetry data to determine machine learning input data based at least in part on at least a portion of the measured temperature values and at least a portion of the setpoint temperature values; and using one or more hardware processors to apply the machine learning input data to a trained anomaly detection machine learning model to determine periodic anomaly metrics. The process provides an automatically determined indication based at least in part on at least a portion of the periodic anomaly metrics.
1 . A method, comprising:
receiving telemetry data associated with one or more refrigeration systems over a first time interval, wherein the telemetry data includes a set of temperature measurements;
receiving anomaly data including an identification of one or more anomaly events associated with the one or more refrigeration systems over the first time interval;
generating a training dataset using the telemetry data;
modifying the training dataset by removing, from the training dataset, a portion of the telemetry data that corresponds to an anomaly event based on the anomaly data; and
training an anomaly detection machine learning model using the modified training dataset, wherein the anomaly detection machine learning model, once trained, is configured to predict an occurrence of an anomaly event occurring over a future time interval.
2 . The method of claim 1 , wherein the telemetry data is collected by one or more sensors associated with the one or more refrigeration systems.
3 . The method of claim 2 , wherein at least one of the one or more sensors is configured to measure an ambient condition external to the one or more refrigeration systems.
4 . The method of claim 1 , wherein the telemetry data is collected periodically and continuously.
5 . The method of claim 1 , wherein generating the training dataset includes modifying the telemetry data using one of forward filling, determined relative values, normalizing values, or linear interpolation.
6 . The method of claim 1 , wherein the anomaly detection machine learning model is a recurrent neural network.
7 . A system, comprising:
one or more processors; and
a non-transitory computer-readable medium comprising instructions that when executed by the one or more processors cause the one or more processors to perform operations including:
receiving telemetry data associated with one or more refrigeration systems over a first time interval, wherein the telemetry data includes a set of temperature measurements;
receiving anomaly data including an identification of one or more anomaly events associated with the one or more refrigeration systems over the first time interval;
generating a training dataset using the telemetry data;
modifying the training dataset by removing, from the training dataset, a portion of the telemetry data that corresponds to an anomaly event based on the anomaly data; and
training an anomaly detection machine learning model using the modified training dataset, wherein the anomaly detection machine learning model, once trained, is configured to predict an occurrence of an anomaly event occurring over a future time interval.
8 . The system of claim 7 , wherein the telemetry data is collected by one or more sensors associated with the one or more refrigeration systems.
9 . The system of claim 8 , wherein at least one of the one or more sensors is configured to measure an ambient condition external to the one or more refrigeration systems.
10 . The system of claim 7 , wherein the telemetry data is collected periodically and continuously.
11 . The system of claim 7 , wherein generating the training dataset includes modifying the telemetry data using one of forward filling, determined relative values, normalizing values, or linear interpolation.
12 . The system of claim 7 , wherein the anomaly detection machine learning model is a recurrent neural network.
13 . A non-transitory computer-readable medium comprising instructions that when executed by one or more processors cause the one or more processors to perform operations including:
receiving telemetry data associated with one or more refrigeration systems over a first time interval, wherein the telemetry data includes a set of temperature measurements;
receiving anomaly data including an identification of one or more anomaly events associated with the one or more refrigeration systems over the first time interval;
generating a training dataset using the telemetry data;
modifying the training dataset by removing, from the training dataset, a portion of the telemetry data that corresponds to an anomaly event based on the anomaly data; and
training an anomaly detection machine learning model using the modified training dataset, wherein the anomaly detection machine learning model, once trained, is configured to predict an occurrence of an anomaly event occurring over a future time interval.
14 . The non-transitory computer-readable medium of claim 13 , wherein the telemetry data is collected by one or more sensors associated with the one or more refrigeration systems.
15 . The non-transitory computer-readable medium of claim 14 , wherein at least one of the one or more sensors is configured to measure an ambient condition external to the one or more refrigeration systems.
16 . The non-transitory computer-readable medium of claim 13 , wherein the telemetry data is collected periodically and continuously.
17 . The non-transitory computer-readable medium of claim 13 , wherein generating the training dataset includes modifying the telemetry data using one of forward filling, determined relative values, normalizing values, or linear interpolation.
18 . The non-transitory computer-readable medium of claim 13 , wherein the anomaly detection machine learning model is a recurrent neural network.