Generator Parameter Estimation Using EKF and PMU Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring generator parameters in power distribution systems require the generator to be taken offline, which is undesirable for real-time monitoring and control.
Innovation Solution
Extended Kalman filtering (EKF) is used with data from phasor measurement units (PMUs to estimate dynamic states and parameters like rotor angle, speed, inertia, and reactance in real-time, employing model decoupling techniques to reduce computational load and enable faster sampling rates for dynamic state estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If generator parameters are measured using conventional methods, then measurement accuracy is achieved, but the generator must be taken offline causing loss of time and productivity
Solution Approach 1:
The generator performs self-measurement by utilizing its own operational data from PMUs to estimate its parameters through the EKF algorithm, eliminating the need for external offline testing and allowing continuous operation while maintaining measurement accuracy
Solution Approach 2:
The patent replaces physical offline testing mechanisms with a computational estimation system using EKF that processes electrical signals from PMUs to derive parameter values, substituting mechanical/test-based measurement with signal-processing-based estimation
2Productivity
If real-time parameter estimation is implemented using EKF, then continuous monitoring capability is improved, but computational complexity increases
Solution Approach 1:
The EKF algorithm is segmented into distinct prediction and update phases that process data in discrete time steps, allowing real-time computation by breaking down the continuous estimation problem into manageable computational segments that can be executed at each sampling interval
Solution Approach 2:
The estimation system is made dynamic by continuously adapting the state estimates and covariance matrices at each time step based on new measurements, allowing the computational process to adjust to changing system conditions in real-time while maintaining manageable complexity through recursive updates
3Measurement precision
If high sampling rates are used for dynamic state estimation, then estimation accuracy is improved, but computational burden increases
Solution Approach 1:
The patent extracts and utilizes readily available PMU measurement data (voltage, current, power flows) that is already being collected for other purposes, eliminating the need for additional specialized sensors and reducing the computational burden by working with existing high-quality measurement data at standard PMU sampling rates
Data Source
AI summary
Various examples are provided for parameter estimation of generators. In one example, among others, a method includes collecting data corresponding to a generator using a phasor management unit (PMU) and estimating dynamic parameters of the generator using extended Kalman filtering (EKF) and the collected PMU data. In another example, a system includes at least one application executable in a processing device that obtains operational data corresponding to a generator and estimates a dynamic parameter of the generator using EKF and the operational data. In another example, an EKF estimator includes a dynamics estimator configured to estimate a state variable of a generator, a geometry estimator configured to estimate phasor values associated with the generator, and a Kalman filter gain configured to determine a correction to the state variable.