Current Transformer Saturation Correction in IEDs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Transformer (CT) saturation during high electrical current faults leads to incorrect current representation, affecting protection and control applications in electrical networks, as existing methods for detection and correction are either computationally intensive or not suitable for low-end Intelligent Electronic Devices (IEDs with limited computational power and sample rates.
Innovation Solution
A method for real-time detection and correction of CT saturation using regression and wavelet filtering in IEDs, which involves estimating errors, applying dynamic correction factors, and selecting corrected sampled values based on thresholds to regenerate accurate primary current measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CT saturation correction methods using neural networks or Auto-Regressive models are applied, then measurement precision is improved, but device complexity and computational burden increase making them unsuitable for low-end IEDs
Solution Approach 1:
The patent replaces complex, expensive correction algorithms (neural networks, AR models) with a simple, computationally lightweight method using basic signal processing operations. The correction approach uses readily available CT output data and applies straightforward calculations that can be executed on low-end IEDs with limited computational resources, effectively substituting expensive computational methods with inexpensive alternatives.
Solution Approach 2:
The patent transforms the correction problem by changing the approach from model-based correction (requiring training data and complex parameters) to a direct signal processing method. By modifying how the correction is computed - using simple arithmetic operations on the CT output signal rather than complex mathematical models - the solution adapts to the constraints of low-end devices while maintaining correction effectiveness.
2Measurement precision
If complex correction algorithms are used, then measurement precision is improved, but productivity and real-time processing capability deteriorate due to computational intensity
Solution Approach 1:
The patent employs computationally inexpensive operations that can be executed rapidly on low-power processors. By using basic arithmetic and simple signal processing instead of intensive algorithms, the system achieves both high measurement precision and fast real-time processing speeds suitable for protective relaying applications where rapid response is critical.
Solution Approach 2:
The patent applies correction only to the specific portions of the signal where saturation occurs, rather than processing the entire signal through complex algorithms. This selective correction approach reduces computational burden while maintaining accuracy where it matters most, enabling real-time processing on resource-constrained devices.
3Measurement precision
If neural network-based saturation detection is implemented, then detection accuracy is improved, but loss of time increases due to extensive training data requirements
Solution Approach 1:
The patent eliminates the need for time-consuming neural network training by using a direct, rule-based detection method. The system uses simple threshold comparisons and signal analysis techniques that provide accurate saturation detection without requiring extensive training datasets or iterative optimization processes, enabling immediate deployment without training time delays.
Solution Approach 2:
The patent incorporates saturation detection capabilities that are immediately available upon system initialization, without requiring preliminary training phases. The detection algorithm is pre-configured with fixed parameters and rules that can be applied directly to incoming CT signals, eliminating the need for offline training and enabling real-time saturation detection from the start of operation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for generating corrected measured current in an Intelligent Electronic Device (IED) that is a true representation of primary current in an electrical network is disclosed. The method uses regression and a first threshold on measured sampled values from CT to detect a deviation instance that indicates a possibility of saturation. Wavelet and second threshold based detection of instances of saturation is then done. Then a regression based correction that uses a dynamic correction factor is implemented in real time to obtain corrected sampled values i.e. corrected measured current. Ending of correction is done based upon a predetermined selection criterion, The generated corrected measured current is used for protection and control functions in the IED. A CT output re-generation module as a functional module in the IED for implementing the method as described above is also disclosed.