Switchgear Thermal Monitoring With Obscuration-Resistant Hotspot Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring switchgear health using infra-red cameras are limited by obscuring elements that reduce detection accuracy and require manual expert analysis, making them costly and inefficient for widespread implementation.
Innovation Solution
An apparatus comprising an input unit, processing unit, and output unit that utilizes a machine learning classifier algorithm, trained on modified infra-red images to remove obscuration effects, enabling accurate detection of hotspots in switchgear and electrical components without human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual expert analysis of thermal images is used to detect hotspots, then detection accuracy is improved, but cost and time consumption increase significantly
Solution Approach 1:
The system performs preliminary processing of thermal images by training a machine learning algorithm to recognize hotspot patterns and obscuring elements. The algorithm is pre-trained on labeled datasets containing various switchgear configurations, enabling automated detection without requiring manual expert analysis for each new image, thus reducing inspection time while maintaining accuracy
Solution Approach 2:
A machine learning algorithm acts as an intermediary between the thermal imaging system and human experts. The algorithm processes thermal images, identifies hotspots, and flags suspicious areas for expert review, reducing the time required for manual inspection while preserving detection accuracy through automated pre-analysis
2Duration of action of stationary object
If infra-red cameras are used to monitor switchgear, then continuous monitoring capability is improved, but obscuring elements reduce detection accuracy
Solution Approach 1:
The system extracts and removes the effects of obscuring elements from thermal images through image processing techniques. By identifying and eliminating the negative impact of caps and other obscuring objects, the system maintains continuous monitoring capability while restoring detection accuracy in previously obscured areas
Solution Approach 2:
The system creates modified versions of thermal images with obscuring element effects removed or corrected. These processed image copies are then analyzed by the machine learning algorithm, allowing continuous monitoring to maintain high detection accuracy even when original images contain obscuring elements
3Ease of manufacture
If a common monitoring solution is applied to all switchgear, then implementation simplicity is improved, but detection accuracy varies due to different switchgear types and geometries
Solution Approach 1:
The machine learning algorithm is designed with universal applicability to handle multiple switchgear types and geometries. By training on diverse datasets encompassing various switchgear configurations, the single algorithm can accurately detect hotspots across different equipment types without requiring type-specific solutions, thus maintaining implementation simplicity while ensuring consistent detection accuracy
Solution Approach 2:
The system adapts to different switchgear types by adjusting algorithm parameters based on the specific equipment being monitored. The machine learning model can modify its detection thresholds and processing parameters according to the geometry and characteristics of each switchgear type, maintaining a common implementation framework while achieving accurate detection across diverse equipment
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 solution enhances the ability to detect hotspots in switchgear and electrical components by automatically removing obscuration effects from infra-red images, providing precise structural health assessments and reducing maintenance costs through automated analysis.
Implementation Method 1
use infra-red cameras to capture thermal variation images
Data Source
AI summary
An apparatus for monitoring a switchgear includes: an input unit; a processing unit; and an output unit. The input unit is provides the processing unit with a monitor infra-red image of the switchgear. The processing unit implements a machine learning classifier algorithm to analyse the monitor infra-red image and determine if there is one or more anomalous hot spots in the switchgear. The machine learning classifier algorithm has been trained based on a plurality of different training infra-red images. The plurality of training infra-red images include a plurality of modified infra-red images generated from a corresponding plurality of infra-red images, each of the modified infra-red images having been modified to remove an effect of obscuration in the image. The output unit outputs information relating to the one or more anomalous hot spots.