Flare Image Analytics for Smoke, Steam, and Flame Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current flare monitoring systems in industrial plants rely on human operators, which are prone to errors due to lapses in judgment and attention, and struggle to accurately distinguish between smoke, steam, and flame, especially in low visibility conditions such as harsh weather or nighttime, leading to potential non-compliance with environmental regulations.
Innovation Solution
A machine deep learning-based self-adaptive industrial automation system that uses real-time images from cameras and plant data to analyze and categorize pixels as smoke, flame, or steam, issuing alerts when conditions fall outside predetermined ranges, employing an optical flow subsystem and data augmentation to enhance accuracy under challenging conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If human operators are used to monitor flare operations continuously, then compliance with EPA regulations can be maintained, but errors occur due to lapses in human judgment and attention
Solution Approach 1:
The patent replaces the mechanical human monitoring system with an automated image processing and analysis system. Cameras capture images of the flare stack, and computer algorithms automatically analyze these images to detect smoke, steam, and flame conditions, eliminating human error while maintaining continuous monitoring capability
Solution Approach 2:
The system performs self-monitoring and self-analysis through automated image processing. The computer system automatically captures, processes, and analyzes flare stack images without requiring human intervention, issuing alerts when abnormal conditions are detected based on pre-programmed criteria
2Ease of operation
If human operators monitor flare operations, then continuous monitoring is possible, but accurate distinction between smoke, steam, and flame is difficult especially in low visibility conditions
Solution Approach 1:
The system uses color analysis and image processing techniques to distinguish between smoke, steam, and flame based on their visual characteristics. The computer algorithms analyze color patterns, intensity, and distribution in captured images to accurately categorize different emissions even in challenging lighting conditions
Solution Approach 2:
The automated image analysis system replaces human visual assessment with computer-based image processing algorithms that can consistently identify and categorize emissions types based on their optical properties, providing more reliable differentiation than human operators
3Reliability
If automated systems are implemented to eliminate human error, then monitoring reliability improves, but system complexity increases
Solution Approach 1:
The system uses a multi-functional integrated platform that combines camera hardware, image processing software, data analysis algorithms, and alert notification capabilities in a single automated system. This universal system handles multiple tasks including image capture, processing, analysis, and communication, reducing overall system complexity despite the advanced capabilities required
Data Source
AI summary
A method and system for advanced flare analytics in a flare operation monitoring and control system is disclosed that contains a data acquisition and augmentation mechanism whereby data is aquired through a plant network including images of the flare operations from single or multi-camera hubs. A machine learning-based self-adaptive industrial automation system process the images and data and assigns pixels to the images according to categories selected from smoke, flame and steam. The results of the analysis are displayed and a notice is issued when the percentage of pixles in a specific category falls outside a predeterminmed range.