US-12625016-B2 - Continuous calibration of sensors in a remotely monitored cooling system

US12625016B2US 12625016 B2US12625016 B2US 12625016B2US-12625016-B2

Abstract

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.

Inventors

Richard Kriss

Assignees

KLATU NETWORKS, INC.

Dates

Publication Date: 20260512
Application Date: 20200918

Claims (20)

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.

Description

PRIORITY This application claims priority to and the benefit of U.S. Provisional Patent Application Ser. No. 62/902,849 filed in the U.S. Patent Office on Sep. 19, 2019 and of U.S. Provisional Patent Application Ser. No. 62/948,292 filed in the U.S. Patent Office on Dec. 15, 2019, the entire content of these applications being incorporated herein by reference as if fully set forth below in its entirety and for all applicable purposes. TECHNICAL FIELD The present invention relates generally to management and calibration of cooling systems, including HVAC, refrigeration and other environmental control systems. BACKGROUND Refrigeration cooling, cell-incubation and heating, ventilation, and air conditioning (HVAC) systems, collectively herein referred as temperature-controlled assets or assets, suffer a loss of operating efficiency over time due to manufacturing defects, mechanical degradation, poor power-quality, adverse environmental factors, deferred maintenance or simple misconfiguration. A loss of reliability or efficiency must be detectable, measurable and correctable, so as to avoid damage to equipment or spoilage to contents. Today however, the systems and methods for determining the state-of-health for a refrigeration cooling, heating and HVAC system follow reactive fail-and-fix procedures, whereby repairs are applied only after the equipment fails—essentially, the failure of the asset is the first evidence of a needed repair. This approach results in decreasing reliability over time for assets that have not yet failed, leading to the highest possible labor and repair costs for assets when they fail in terms of energy, repair and maintenance costs. In Life Science and Pharmaceutical applications as just one example, high-reliability, uniformity and precision control of temperature is also required to protect and assure the quality of research, manufacturing, transportation logistics or the storage of products and commodities contained within a temperature-controlled asset, shipping container or in a climate-controlled room. Products and commodities must be stored at prescribed temperatures which are often governed by government regulation and subject to audit. Compliance requires that the stability and the uniformity of temperatures in the cabinet are verified which entails the gathering and analysis of temperature measurements from multiple points, before products and commodities are added to the asset and, periodically thereafter according to standard operating procedures—typically every six (6) or twelve (12) months. The term applied to the inspection protocols which demonstrate compliance is called validation. The validation process is labor intensive and expensive, costing $2,000 to $5,000 per asset or room and can take several days. Before the validation process can begin, the contents of the temperature chamber must be removed and to assess temperature stability and uniformity, a number of sensors, sometimes a dozen or more are placed on or about the shelves or walls of the chamber or room to measure temperatures from top-to-bottom, side-to-side and front-to-back. After the validation protocol is completed, the sensors are removed, and the asset is deemed to be validated if a test protocol is completed and no deviations or exceptions are noted. The sensors and test equipment are then moved to the next asset and the validation process is repeated. In a typical Life Science facility with 300 Ultra-Low Temperature Freezers, the cost of a validation project can exceed $600,000 per year. In addition to the costs to administer a validation, the process can also induce and operational issues due to scheduling delays, lack of skills or the availability of specialized equipment. Present industry best practices which rely on scheduled maintenance or validations are inefficient because an entire population of refrigeration assets must be inspected even though only a percentage might require repair, maintenance or validation. Equipment malfunctions, mechanical degradation, deferred maintenance and environmental conditions can cause temperatures to fall outside allowable limits (referenced by FDA as “exceptions”). When they occur, multi-million-dollar product losses are possible and are not uncommon. Continuous real-time monitoring systems are expensive and uncommon in the industry today. To avoid the uncertainties of maintaining equipment in a validated state, some operators adopt costly mitigation strategies involving the replacement of refrigeration assets with a 10-year life after only five years of service, even though there may be nothing wrong with some refrigeration assets. One example involves ultra-low temperature freezers (UL freezers or −80 C freezers)—an estimated 500,000 of which are deployed in Life Science and Pharmaceutical manufacturing applications. Each ULT consumes the equivalent energy of an entire house, can account for 25-30% of all electricity consumed within a Life Science facility,