PMU Phasor Estimation With Recursive DFT Frequency Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing PMU systems fail to provide accurate phasor measurements due to computational errors in DFT algorithms, sensor inaccuracies, and non-compliance with IEEE standards, especially at medium or high voltage levels, limiting their use in distribution circuits for protection and diagnostics.
Innovation Solution
Implementing a system that corrects sensor signals, employs parallel recursive DFT algorithms with variable window lengths, linear interpolation, and frequency filtering to improve frequency estimation and RoCoF calculation, while accounting for sampling rate deviations and sensor frequency dependencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional DFT algorithms are used in PMU systems, then the computational process is simple, but the measurement precision of phasor measurements deteriorates due to computational errors
Solution Approach 1:
The patent divides the DFT computation into two separate algorithms running in parallel: a first recursive DFT algorithm and a second recursive DFT algorithm. Each algorithm processes the input signal independently with different initialization and update equations, and their results are combined to produce the final phasor measurement. This segmentation allows each algorithm to have simplified computational steps while collectively achieving higher measurement precision by reducing computational errors.
2Measurement precision
If sensor signals are used directly without correction, then the device complexity is low, but the measurement precision deteriorates due to sensor inaccuracies and frequency dependencies
Solution Approach 1:
The patent implements frequency estimation and correction mechanisms that use feedback from the measured signal characteristics. The system estimates the actual frequency of the input signal, compares it with the nominal frequency, and applies correction factors to compensate for sensor frequency dependencies and sampling rate deviations. This feedback-based correction improves measurement precision by accounting for real-world sensor inaccuracies without requiring complex hardware modifications.
3Productivity
If DFT algorithms are executed frequently to capture real-time dynamics, then the productivity is high, but the measurement precision deteriorates due to computational errors accumulating
Solution Approach 1:
The patent employs recursive DFT algorithms that continuously update phasor measurements using the current input signal and previous computation results. Instead of performing complete DFT computations at each time step, the recursive algorithms maintain running sums and updates, allowing continuous real-time measurements with reduced computational error accumulation. The parallel structure ensures that one algorithm can be reset while the other continues, maintaining continuous useful action without precision loss.
4Measurement precision
If fixed window length DFT is used, then the ease of operation is high, but the measurement precision deteriorates when frequency variations occur
Solution Approach 1:
The patent implements variable window length DFT by allowing the DFT window to adapt its length based on the estimated frequency of the input signal. When frequency variations are detected, the window length is adjusted to maintain an integer number of cycles within the analysis window. This dynamic adjustment improves frequency estimation accuracy and phasor measurement precision during frequency transients, while the adaptation logic remains relatively simple to implement.
Data Source
AI summary
A system and method for generating phasors having a phase angle and magnitude from a sinusoidal power signal, and using the phasors to generate a frequency and RoCoF of the power signal. The system and method provide improvements over known systems and methods by correcting the frequency of sensor signals, providing a simplified phasor correction equation, reducing errors in a non-recursive DFT algorithm, improving frequency estimation by spacing out DFTs, correcting the sampling rate, providing linear interpolation of frequency and/or providing a variable DFT window length.