OGI Camera Robot Inspection With Pre-Cooling and Leak Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robots are not well-suited for detecting gas leaks at facilities, as they often lack appropriate sensors, generate excessive heat impairing sensor operation, and capture large volumes of video data without distinguishing relevant features for gas leak detection.
Innovation Solution
A robot system equipped with a sensor system that obtains inspection path information, adjusts camera orientations, and uses an optical gas imaging camera with a cooling system to detect gas leaks, while employing machine learning for image classification and data management to focus on relevant data capture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an OGI camera with refrigeration system is used to detect gas leaks, then detection capability is improved, but device complexity increases
Solution Approach 1:
The refrigeration system is activated in advance before the robot reaches inspection locations to pre-cool the OGI camera sensor. This preliminary cooling ensures the sensor is at optimal temperature for gas leak detection when needed, rather than attempting to cool it rapidly at the moment of inspection.
Solution Approach 2:
The refrigeration system operates periodically - cooling the sensor when the robot is traveling between locations and pausing when the robot is stationary at inspection points. This periodic operation reduces energy consumption and thermal stress while maintaining detection capability.
2Area of stationary object
If the robot travels between locations, then inspection coverage is improved, but energy consumption increases due to continuous refrigeration
Solution Approach 1:
The refrigeration system is controlled to operate only during specific periods - when the robot is traveling between inspection locations. The system pauses cooling when the robot is stationary at inspection points, thereby reducing overall energy consumption while maintaining the ability to detect gas leaks at all required locations.
Solution Approach 2:
The refrigeration system activates in advance during transit to pre-cool the sensor before the robot arrives at inspection locations. This timing strategy ensures the sensor is ready for detection without requiring continuous cooling operation.
3Loss of information
If the robot captures video data at all locations, then data completeness is improved, but data processing burden increases
Solution Approach 1:
The system extracts and captures only the specific video frames or image data that contain relevant gas leak detection information, rather than storing all video data. The OGI camera is configured to capture images only when gas leaks are detected or at predetermined critical locations, filtering out unnecessary data during the inspection process.
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
The system effectively detects gas leaks by optimizing sensor operation, reducing unnecessary data capture, and utilizing machine learning for accurate classification, enhancing the efficiency and accuracy of gas leak inspection.
Implementation Method 1
A robot system equipped with a sensor system that obtains inspection path information, adjusts camera orientations, and uses an optical gas imaging camera with a cooling system to detect gas leaks
Implementation Method 2
uses an optical gas imaging camera with a cooling system to detect gas leaks
Data Source
AI summary
Provided is a process including: obtaining a first set of one or more images generated via an optical gas imaging (OGI) camera and a second set of one or more images generated via a second camera, wherein the first set and the second set correspond to a first location of a plurality of locations in a facility; classifying, via a convolutional neural network (CNN), the first set of one or more images according to whether the one or more images depict a gas leak; and storing a result of classifying the first set in memory.