Switchgear Overheating Detection Using Temperature-Current Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing monitoring systems for switchgear fail to provide early warnings of developing faults and do not detail the cause of overheating issues, relying on temperature thresholds that do not account for current load variations.
Innovation Solution
A system utilizing a discriminator of a trained conditional generative adversarial network (CGAN) to analyze temperature and current data from multiple locations, determining the probability of normalcy based on healthy data, raising alarms for abnormal conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If temperature threshold monitoring is used, then simple alarm detection is achieved, but early warning capability and fault cause identification are lost
Solution Approach 1:
The patent segments the monitoring system into multiple functional modules: temperature sensors at different locations, current sensors, a processing unit that analyzes the relationship between temperature and current, and an output unit. This segmentation allows the system to maintain simple alarm detection while adding analytical capabilities to identify fault causes through the correlation analysis between temperature and current data.
Solution Approach 2:
The patent introduces an intermediary processing unit that acts as a mediator between the sensors and the alarm system. This processing unit analyzes the relationship between temperature and current data, determining whether temperature increases are due to load changes or actual faults. This intermediary layer preserves the simplicity of alarm detection while adding the capability to identify fault causes through correlation analysis.
2Ease of manufacture
If fixed temperature thresholds are used, then implementation simplicity is maintained, but accuracy in detecting actual faults deteriorates due to current load variations
Solution Approach 1:
The patent transforms the static temperature threshold into a dynamic evaluation process. Instead of using fixed thresholds, the processing unit dynamically evaluates whether temperature increases are abnormal by analyzing the relationship between temperature and current data. The system determines if temperature rises are proportional to current increases (normal) or exceed expected proportions (abnormal), adapting the assessment criteria based on actual operating conditions.
Solution Approach 2:
The patent changes the monitoring parameter from absolute temperature values to the relationship between temperature and current. By analyzing how temperature changes relative to current changes, the system can distinguish between temperature increases caused by normal load variations and those caused by actual faults. This parameter transformation maintains implementation simplicity while significantly improving detection accuracy.
3Reliability
If multiple sensors and complex analysis are implemented, then early detection and fault identification capability are improved, but system complexity increases
Solution Approach 1:
The patent implements a universal processing unit that performs multiple functions: acquiring temperature and current data, analyzing the relationship between these parameters, determining whether temperature increases are abnormal, and controlling alarm output. This multi-functional approach allows the system to achieve early fault detection and identification capabilities without proportionally increasing system complexity, as a single processing unit handles all analytical tasks.
Solution Approach 2:
The patent incorporates feedback mechanisms where the processing unit continuously monitors the relationship between temperature and current, comparing actual temperature increases against expected increases based on current changes. This feedback loop enables the system to automatically distinguish between normal and abnormal conditions, providing early fault detection without requiring complex external control systems or multiple independent analysis modules.
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
Enables early detection of overheating anomalies by assessing temperature values against current loads, providing detailed information on fault causes and preventing catastrophic failures.
Implementation Method 1
The at least one temperature sensor is configured to acquire temperature data for at least one location of a switchgear
Implementation Method 2
The at least one current sensor is configured to acquire electrical current data for the at least one location of the switchgear
Data Source
Figure 1~2
Figure 3~6
AI summary
The present invention relates to a system (10) for monitoring a switchgear, the system comprising: - at least one temperature sensor (20); - at least one current sensor (30); - a processing unit (40); and - an output unit (50). The at least one temperature sensor is configured to acquire temperature data for at least one location of a switchgear (60). The at least one temperature sensor is configured to provide the temperature data to the processing unit. The at least one current sensor is configured to acquire electrical current data for the at least one location of the switchgear. The at least one current sensor is configured to provide the electrical current data to the processing unit. The processing unit is configured to implement a discriminator (70) of a trained conditional generative adversarial network, CGAN, wherein the CGAN comprises a generator and the discriminator. The processing unit is configured to determine a state of the switchgear, wherein the determination of the state of the switchgear comprises utilization of the discriminator without utilization of the generator, and wherein the determination of the state of the switchgear comprises utilization of the discriminator to analyze temperature data for one or more locations of the at least one location and electrical current data for the one or more locations. The output unit is configured to output the determined state of the switchgear when the determined state of the switchgear is determined not to be normal or not to be healthy.