Current Transformer Saturation Detection via Pre-computed Model Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting current transformer saturation are prone to errors and resource-intensive, requiring extrapolation of sample values and threshold comparisons at each time step.
Innovation Solution
Determine running sums and variances of sample values to quickly and reliably detect saturation by comparing calculation parameters with default values, reducing resource expenditure and allowing for efficient storage and reuse of model parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extrapolation of sample values and threshold comparisons are performed at each time step, then saturation detection accuracy is improved, but computational resource expenditure increases
Solution Approach 1:
The patent pre-calculates and stores model parameters (amplitude, frequency, phase) during normal operation before saturation occurs. When saturation is detected, these pre-computed parameters enable rapid correction without requiring resource-intensive real-time extrapolation, thus resolving the contradiction between detection accuracy and computational resource usage
Solution Approach 2:
The patent creates a corrected sample signal that copies the structure and characteristics of the original signal but uses pre-determined model parameters instead of performing full extrapolation at each time step. This copying approach maintains detection accuracy while significantly reducing computational resources required during saturation events
2Use of energy by moving object
If model parameters are stored and reused, then resource consumption is reduced, but detection reliability may be compromised
Solution Approach 1:
The patent implements feedback mechanisms where model parameters are continuously monitored and updated based on signal characteristics. The system checks whether stored parameters remain valid by comparing against current signal conditions, and automatically recalibrates when drift is detected, thus maintaining reliability while benefiting from parameter reuse
Solution Approach 2:
The patent makes the parameter storage system dynamic by implementing validity checks and conditional updates. Model parameters are stored during normal operation but can be refreshed when saturation patterns change or signal characteristics drift, ensuring the system adapts to changing conditions while maintaining the efficiency benefits of parameter caching
Data Source
Figure 1
Figure 2
AI summary
Method for saturation detection of a current transformer, wherein the converted current signal is sampled discretely in time and the sample value (8) is stored in a measurement memory by a computing unit for each time step, the model parameters of a signal model having an AC and an exponential DC element are determined on the basis of several sample values (8) and the model parameters are stored in a parameter memory.In order to design a method of the type described above in such a way that the error susceptibility of a current transformer saturation detection is reduced despite lower resource expenditure in normal operation, it is proposed that in successive time steps a running sum of the sampled values (8), the RMS value of the AC element with its mean and variance, the base of the exponential DC element with its mean, and the coefficient of the DC element with its running sum are determined as computational parameters, and a saturation signal is output if the mean of the base lies outside a base standard range, the mean of the RMS value lies above a first nominal current limit, the variance of the RMS value lies above a minimum variance, and the sign of the sampled value (8) matches the sign of the running sum of the sampled values (8) and the running sum of the coefficients of the DC element.