Anomaly detection for refrigeration systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional maintenance techniques for refrigeration systems often result in unnecessary expenses and lost productivity due to late detection of failures or over-maintenance, as they fail to accurately account for the nuances of refrigeration systems and are prone to false positives.
Innovation Solution
Anomaly detection using a machine learning model trained with telemetry data from refrigeration systems, including temperature values and setpoint temperatures, to predict equipment failures with high confidence, allowing for more precise scheduling of maintenance and reducing unnecessary service trips.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional condition-based maintenance schedules are used, then maintenance is performed based on system health assessment, but false positives increase and accuracy decreases
Solution Approach 1:
The patent transforms maintenance scheduling from conventional condition-based approaches to anomaly-based scheduling by changing the detection parameters. Instead of using generic health assessments, the system uses trained machine learning models that analyze multiple telemetry parameters (temperature, pressure, power consumption, runtime hours) to detect anomalies with higher precision, thereby reducing false positives while maintaining reliable maintenance scheduling.
2Loss of energy
If reactive maintenance is performed, then expenses are reduced by avoiding unnecessary maintenance, but productivity is lost due to late failure detection
Solution Approach 1:
The system performs preliminary anomaly detection and failure prediction before actual system failures occur. By continuously monitoring telemetry data and comparing it against trained models, the system identifies anomalies early, enabling maintenance to be scheduled proactively rather than reactively. This preliminary action prevents unexpected failures that would cause productivity loss while avoiding unnecessary maintenance expenses.
3Reliability
If planned preventative maintenance is performed according to schedule, then system reliability is maintained, but unnecessary maintenance trips increase expenses
Solution Approach 1:
The patent transitions from static, time-based maintenance scheduling to dynamic anomaly-based scheduling. Instead of performing maintenance at fixed intervals regardless of system condition, the system dynamically adjusts maintenance timing based on real-time anomaly detection results. This allows maintenance to be performed only when anomalies indicate potential failures, maintaining system reliability while eliminating unnecessary maintenance trips and associated expenses.
Data Source
AI summary
In various embodiments, 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.