Refrigerant leak detection using a sensor-reading context analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional refrigerant leak detection systems using threshold-based sensors often generate false alarms when using moderate-to-low global warming potential (GWP) refrigerants, leading to system downtime and mistrust in the detection system due to high false alarm rates.
Innovation Solution
A detection assembly that incorporates a sensor network and a controller performing sensor-reading context analysis, utilizing a machine learning algorithm to classify patterns in sensor outputs within a context time window, including both triggering-sensor and context-sensor data such as refrigerant concentration, ambient temperature, and humidity, to accurately determine refrigerant leaks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If threshold-based detection schemes are used, then the detection system is simple to operate, but the false alarm rate is high
Solution Approach 1:
The system transitions from static threshold-based detection to dynamic pattern recognition. The controller continuously monitors sensor outputs and analyzes patterns of readings over time, adapting to varying operating conditions. This dynamic approach allows the system to distinguish between actual leaks and transient sensor variations, significantly reducing false alarms while maintaining ease of operation.
Solution Approach 2:
The system implements feedback by continuously monitoring sensor outputs and using this information to adjust detection decisions. The controller analyzes historical sensor data and current readings in context, using feedback loops to improve detection accuracy over time. This feedback mechanism enables the system to learn from past readings and reduce false alarm rates while keeping the operation simple for users.
2Measurement precision
If context sensors are added to the sensor network, then the detection accuracy is improved, but the device complexity increases
Solution Approach 1:
The system merges triggering sensors that detect refrigerant presence with context sensors that monitor environmental conditions. By combining these sensor types into a unified sensor network, the system achieves improved detection accuracy through contextual analysis. The controller integrates data from all sensors and analyzes their combined outputs, allowing accurate leak detection while presenting a unified, manageable interface to users.
Solution Approach 2:
The controller serves multiple functions: it processes triggering sensor data, analyzes context sensor readings, performs pattern recognition, and generates detection decisions. This multi-functional approach consolidates complexity into a single intelligent component rather than requiring separate systems for each function. The universal controller handles diverse sensor inputs and produces accurate leak detection, improving measurement precision without proportionally increasing overall device complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Significantly reduces false alarm rates while maintaining detection reliability, minimizing system downtime and technician visits, and enhancing the trustworthiness of refrigerant sensors.
Implementation Method 1
Conventional refrigerant detection assemblies utilize threshold-based refrigerant leak detectors, such as a nondispersive infrared (NDIR) sensor
Implementation Method 2
Conventional refrigerant detection assemblies utilize threshold-based refrigerant leak detectors, such as a nondispersive infrared (NDIR) sensor or a metal-oxide-semiconductor-based (MOS-based) sensor
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A detection assembly (126) operable to detect a refrigerant leak event includes a sensor network (110) and a controller (120). The sensor network is operable to generate sensor outputs (220) including triggering-sensor (TS) outputs and triggering-sensor context (TSC) outputs. The controller is operable to perform a sensor-reading context analysis on the sensor outputs. The sensor-reading context analysis includes accessing a set of the sensor outputs (A, B) that occurred within a context time window (230, 240), along with determining that a pattern of the set of sensor outputs represents the refrigerant leak event.