Multi-Spectral Flame Detection for False Alarm Discrimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing flame detection systems fail to accurately distinguish between actual flames and false flame sources such as arc welding, halogen lights, or heaters due to similar spectral reflections, leading to false alarms.
Innovation Solution
A flame detection system utilizing a combination of image capturing units and flame detectors that capture multiple spectral bands of data, including near and long band IR data and ultraviolet UVC band data, with a processor that classifies and weights these data to determine the presence of flames by comparing weighted counts against threshold values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing flame detection systems use simple spectral detection, then the device complexity is low, but the measurement precision deteriorates due to inability to distinguish actual flames from false sources
Solution Approach 1:
The system segments the detection process into multiple independent components: image capturing unit for visual analysis, flame detector for spectral analysis (divided into UV and IR sensors), and processor for integrated decision-making. Each component handles a specific aspect of flame detection, allowing the system to achieve high precision through coordinated analysis of multiple data streams rather than relying on a single complex sensor.
2Reliability
If the system uses multiple spectral bands and classification algorithms, then the reliability improves by reducing false alarms, but the loss of time increases due to complex data processing
Solution Approach 1:
The system performs preliminary classification of spectral data and images before final flame determination. The processor pre-processes the spectral band data and image data separately, identifying characteristic patterns and assigning weights based on pre-established criteria. This preliminary action allows the system to quickly filter out false alarm sources (such as welding arcs or sunlight) before committing to a final flame detection decision, thereby maintaining high reliability without excessive delay.
3Measurement precision
If the system integrates image capturing and spectral detection, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The system merges image capturing and spectral detection into a unified flame detection process. The processor receives both image data from the image capturing unit and spectral band data from the flame detector simultaneously, then integrates these data streams through a coordinated analysis process. The image data provides visual context while the spectral data provides chemical composition information, and their merger enables the system to achieve high identification accuracy by cross-validating multiple independent measurement modalities.
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
Effectively distinguishes between actual flames and false sources, reducing false alarms by accurately identifying flames through a weighted count and probability-based analysis.
Implementation Method 1
at least one flame detector configured to capture a first set of spectral band data and a second set of spectral band data present in the FOV... the first set of spectral band data corresponds to a near band IR data... the second set of spectral band data corresponds to a long band IR data, a wide band IR data, and an ultraviolet UVC band data
Implementation Method 2
the second set of spectral band data corresponds to a long band IR data, a wide band IR data, and an ultraviolet UVC band data
Data Source
Figure 1
Figure 2
Figure 2
AI summary
A flame detection system is disclosed. The flame detection system comprises a image capturing unit configured to capture images of a field of view (FOV), a flame detector configured to capture a first set of spectral band data and a second set of spectral band data, and a processor configured to classify the one or more images and first set of spectral band data, assign a weight to each second set of spectral band data, receive a flame count for each second set of spectral band data, determine a weighted count for each flame count, determine a quantity of weighted counts exceeding its respective threshold value, and determine that a flame exists within the FOV when the quantity is greater than or equal to a quantity threshold value, and determine that a flame does not exist within the FOV when the quantity is less than the quantity threshold value.