Continuous calibration of sensors in a remotely monitored cooling system
Systems, methods and apparatus may be applicable to managing and monitoring refrigeration assets, including refrigeration plants and cold-storage facilities comprising large numbers of refrigeration assets. A method of managing refrigeration systems includes receiving measurements captured by a plurality of sensors deployed with a refrigeration asset, the measurements being related to temperatures within a temperature-controlled chamber of the refrigeration asset, identifying a difference between a first temperature measurement obtained from measurements provided by a first sensor and a second temperature measurement obtained from measurements provided by at least one sensor, and calibrating the first sensor based on the difference between the first temperature measurement and the second temperature measurement.
1 . A method of managing refrigeration systems, comprising:
receiving temperature measurements that are continuously captured by clusters of sensors, the clusters of sensors being deployed at different locations within a temperature-controlled asset, each cluster of sensors including a plurality of sensors configured to measure temperature at the deployed location of the each cluster of sensors within a compartment of the temperature-controlled asset while the temperature-controlled asset is in operation;
continuously recalibrating each sensor in a first cluster of sensors based on the temperature measurements generated by the first cluster of sensors;
automatically detecting a calibration error based on a difference identified between temperature measurements in a time series of sensor data received from the first cluster of sensors and at least one other time series of sensor data received from the first cluster of sensors; and
recalibrating a first sensor in the first cluster of sensors without manual intervention when differences between temperature cycles reported by the first sensor in the first cluster of sensors and temperature cycles reported by one or more other sensors in the first cluster of sensors vary from baseline differences,
wherein each sensor in the first cluster of sensors is recalibrated during a continuous validation process involving the temperature-controlled asset.
2 . The method of claim 1 , wherein the at least one other time series of sensor data includes temperature measurements previously received from the cluster of sensors under operational conditions.
3 . The method of claim 1 , wherein the at least one other time series of sensor data is obtained from a comparable asset, a peer group of assets, a population of assets or a simulated asset.
4 . The method of claim 3 , further comprising:
determining that the first sensor in the first cluster of sensors is out of calibration when temperature measurements provided by two or more other sensors in the first cluster of sensors are consistent with temperature measurements provided by a second cluster of sensors and inconsistent with temperature measurements provided by the two or more other sensors in the first cluster of sensors.
5 . The method of claim 1 , further comprising:
performing a frequency domain analysis of the temperature measurements captured by the clusters of sensors; and
determining that the temperature measurements captured by the first sensor in the first cluster of sensors deviate from the at least one other time series of sensor data based on the frequency domain analysis.
6 . The method of claim 1 , further comprising:
determining onset of failure of equipment associated with the asset based on a change in stability or distribution of thermal energy within the asset indicated by the clusters of sensors.
7 . The method of claim 1 , wherein an initial calibration of the plurality of sensors is accomplished by:
calibrating the clusters of sensors prior to initial operation; and
calibrating differences in measurements provided by pairs of sensors in each cluster of sensors after calibration.
8 . The method of claim 7 , further comprising:
recalibrating the differences in measurements provided by the pairs of sensors after a change in conditions within the asset.
9 . The method of claim 8 , wherein conditions within the asset are changed when an object is added, moved or removed.
10 . The method of claim 1 , further comprising:
determining a loss of calibration or accuracy of the first sensor in the first cluster of sensors based on a determination that the first sensor has lost correlation or covariance with other sensors in the first cluster of sensors.
11 . The method of claim 1 , further comprising:
using the clusters of sensors to continuously validate the temperature-controlled asset based on an assessment of stability and uniformity of temperatures within a chamber of the temperature-controlled asset.
12 . The method of claim 1 , further comprising:
using a neural network to determine that the temperature measurements captured by the first sensor in the first cluster of sensors deviate from the at least one other time series of sensor data.
13 . The method of claim 1 , further comprising:
using pattern matching to determine that the temperature measurements captured by the first sensor in the first cluster of sensors deviate from the at least one other time series of sensor data.
14 . The method of claim 1 further comprising:
using a neural network to calibrate the clusters of sensors.
15 . The method of claim 1 further comprising:
using pattern matching to calibrate the clusters of sensors.
16 . The method of claim 1 further comprising:
using a neural network to continuously validate the temperature-controlled asset based on measurements received from one or more continuously calibrated clusters of sensors.
17 . The method of claim 1 further comprising:
using pattern matching to continuously validate the temperature-controlled asset based on measurements received from one or more continuously calibrated clusters of sensors.
18 . An apparatus for managing refrigeration systems, comprising:
one or more communication interfaces, including a wireless communication interface configured to couple the apparatus to a wireless communication network;
a sensor interface circuit configured to receive temperature measurements that are continuously captured by clusters of sensors, the clusters of sensors being deployed at different locations within a temperature-controlled asset, each cluster of sensors including a plurality of sensors configured to measure temperature at the deployed location of the each cluster of sensors within a compartment of the temperature-controlled asset while the temperature-controlled asset is in operation; and
a processing circuit configured to:
continuously recalibrate each sensor in a first cluster of sensors based on the temperature measurements generated by the first cluster of sensors;
automatically detect a calibration error based on a difference identified between temperature measurements in a time series of sensor data received from the first cluster of sensors and at least one other time series of sensor data received from the first cluster of sensors; and
recalibrate a first sensor in the first cluster of sensors without manual intervention when differences between temperature cycles reported by the first sensor in the first cluster of sensors and temperature cycles reported by one or more other sensors in the first cluster of sensors vary from baseline differences,
wherein each sensor in the first cluster of sensors is recalibrated during a continuous validation process involving the temperature-controlled asset.
19 . The apparatus of claim 18 , wherein the at least one other time series of sensor data includes temperature measurements previously received from the cluster of sensors under operational conditions.
20 . The apparatus of claim 18 , wherein the at least one other time series of sensor data is obtained from a comparable asset, a peer group of assets, a population of assets or a simulated asset.
21 . The apparatus of claim 20 , wherein the processing circuit is configured to:
determine that the first sensor in the first cluster of sensors is out of calibration when temperature measurements provided by two or more other sensors in the first cluster of sensors are consistent with temperature measurements provided by a second cluster of sensors and inconsistent with temperature measurements provided by the two or more other sensors in the first cluster of sensors.
22 . The apparatus of claim 18 , wherein the processing circuit is configured to:
perform a frequency domain analysis of the temperature measurements captured by the clusters of sensors; and
determine that the temperature measurements captured by the first sensor in the first cluster of sensors deviate from the at least one other time series of sensor data based on the frequency domain analysis.
23 . The apparatus of claim 18 , wherein the processing circuit is configured to:
determine onset of failure of equipment associated with the asset based on a change in stability or distribution of thermal energy within the asset indicated by the clusters of sensors.
24 . The apparatus of claim 18 , wherein the processing circuit is configured to:
calibrate the clusters of sensors prior to initial operation; and
calibrate differences in measurements provided by pairs of sensors in each cluster of sensors after calibration.
25 . The apparatus of claim 24 , wherein the processing circuit is configured to:
recalibrating the differences in measurements provided by the pairs of sensors after a change in conditions within the asset.
26 . The apparatus of claim 25 , wherein conditions within the asset are changed when an object is added, removed or moved.
27 . The apparatus of claim 18 , wherein the processing circuit is further configured to:
use the clusters of sensors to continuously validate the temperature-controlled asset based on an assessment of stability and uniformity of temperatures within a chamber of the temperature-controlled asset, wherein the temperature-controlled asset is continuously validated in accordance with an industrial standard or government regulation.
28 . A non-transitory processor-readable storage medium configured with code that, when executed by a processor, causes the processor to:
receive temperature measurements that are continuously captured by clusters of sensors, the clusters of sensors being deployed at different locations within a temperature-controlled asset, each cluster of sensors including a plurality of sensors configured to measure temperature at the deployed location of the each cluster of sensors within a compartment of the temperature-controlled asset while the temperature-controlled asset is in operation;
continuously recalibrate each sensor in a first cluster of sensors based on the temperature measurements generated by the first cluster of sensors;
automatically detect a calibration error based on a difference identified between temperature measurements in a time series of sensor data received from the first cluster of sensors and at least one other time series of sensor data received from the first cluster of sensors; and
recalibrate a first sensor in the first cluster of sensors without manual intervention when differences between temperature cycles reported by the first sensor in the first cluster of sensors and temperature cycles reported by one or more other sensors in the first cluster of sensors vary from baseline differences,
wherein each sensor in the first cluster of sensors is recalibrated during a continuous validation process involving the temperature-controlled asset.