Power Grid Parameter Error Detection Using Sparse Matrix Methods
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional state estimation systems in power grids fail to accurately detect and distinguish network parameter errors from measurement errors, especially in large utility power grids, due to computational inefficiencies and inability to handle simultaneous errors.
Innovation Solution
The implementation of a computationally efficient process using sparse inverse methods and Lagrange multipliers to identify network parameter errors, which reduces CPU time and memory requirements by orders of magnitude, allowing for simultaneous inspection of all network parameters without requiring a suspect set selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional state estimation systems are used to detect network parameter errors, then measurement errors can be identified, but the systems fail to distinguish parameter errors from measurement errors and cannot scale to large utility power grids
Solution Approach 1:
The patent segments the error detection problem into two distinct components: measurement error detection (using normalized residuals) and parameter error detection (using Lagrange multipliers). This segmentation allows the system to differentiate between the two error types and apply appropriate detection methods for each, enabling accurate error identification while maintaining system scalability to large power grids.
2Reliability
If comprehensive inspection of all network parameters is performed, then parameter errors can be detected, but computational burden increases significantly
Solution Approach 1:
The patent extracts and utilizes only the necessary components for parameter error detection from the full state estimation calculation. By computing Lagrange multipliers associated with network parameters and comparing them against thresholds, the system detects parameter errors without performing comprehensive inspection of all parameters, thereby reducing computational burden while maintaining detection reliability.
3Measurement precision
If full covariance matrix computation is performed for error detection, then accurate error identification is achieved, but memory requirements increase by orders of magnitude
Solution Approach 1:
The patent extracts and computes only the diagonal elements of the covariance matrix, which are sufficient for calculating the variance of Lagrange multipliers and performing error detection. This extraction approach maintains accurate error identification by preserving the necessary statistical information while reducing memory requirements by orders of magnitude compared to computing the full covariance matrix.
4Productivity
If suspect set selection is required for parameter error detection, then computational complexity is reduced, but the ability to detect errors outside the suspect set is lost
Solution Approach 1:
The patent implements a universal error detection mechanism using Lagrange multipliers that can detect parameter errors throughout the entire network without requiring suspect set selection. The method computes Lagrange multipliers for all network parameters and compares them against thresholds, providing both broad error detection coverage and computational efficiency through the use of sparse matrix techniques.
Data Source
AI summary
Techniques and apparatus for parameter error detection in a power system based on state estimation are described. In one embodiment, for example, an efficient process may be used to derive and compute only the necessary subset of the gain matrix and covariance matrix, thus avoiding the computation and storage of large dense matrices. The described efficient process can be applied either to single-scan or multiple-scan schemes. Other embodiments are described.