Transformer Fault Detection Using 2D Feature Space Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Power transformers in drive systems can develop inter-turn short-circuit faults, leading to unexpected shutdowns and extended maintenance, which existing technologies fail to detect efficiently, resulting in reduced operational availability and reliability.
Innovation Solution
A method and apparatus using a fault detection model to generate a transition region in a two-dimensional feature space, determining feature indicators for a power transformer, and predicting an inter-turn short fault by identifying feature indicators within a fault region, thereby minimizing false positives and efficiently detecting faults.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing fault detection technologies are used for power transformers, then the detection process is simple, but the detection accuracy is low resulting in false positives and missed faults
Solution Approach 1:
The patent transitions from traditional single-threshold detection to a two-dimensional feature space approach. By plotting feature indicators (such as ratio of differential current to restraining current versus rate of change of these features) on a 2D plane and defining regions (healthy, transition, fault), the system achieves more accurate fault detection. This dimensional expansion allows better discrimination between normal variations and actual faults, reducing false positives while maintaining manageable system complexity through systematic region-based decision rules.
2Reliability
If traditional fault detection methods are applied, then the system operation is continuous, but unexpected shutdowns occur due to undetected inter-turn short faults
Solution Approach 1:
The patent implements preliminary fault detection by continuously monitoring feature indicators and comparing them against predefined regions in the 2D feature space before catastrophic failures occur. The transition region concept allows the system to detect early signs of inter-turn short faults (such as incipient winding defects) before they lead to complete transformer failure. This preliminary detection capability enables proactive maintenance scheduling, preventing unexpected shutdowns and improving operational availability without requiring overly complex detection equipment.
3Reliability
If extensive maintenance is performed to ensure reliability, then the transformer remains operational, but downtime and maintenance costs increase
Solution Approach 1:
The patent establishes a feedback mechanism where feature indicators are continuously measured, plotted in the 2D feature space, and compared against healthy/transition/fault regions. This closed-loop monitoring provides real-time feedback on transformer health status, allowing operators to distinguish between normal operational variations (remaining in healthy region) and actual degradation trends (moving toward transition or fault regions). This targeted approach enables maintenance to be performed only when necessary, based on actual measured conditions rather than fixed schedules, thereby reducing unnecessary downtime and maintenance costs while maintaining reliability.
Data Source
AI summary
For transformer fault detection, a method generates a transition region that separates a health region and a fault region in a two-dimensional feature space of two feature indicators for a plurality of operation conditions using a fault detection model for a power transformer type. The method determines the feature indicators of a given power transformer of the power transformer type. The method determines whether the feature indicators in the transition region satisfy a fault condition. The method predicts an inter-turn short fault for the given power transformer in response to satisfying the fault condition or the feature indicators being in the fault region.