Dynamic Process Estimation for Weakly Observable Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating the dynamic process of nodes without Phasor Measurement Units (PMU) in electric power systems are inaccurate and inefficient, relying on recalculated Jacobian matrices and grid parameters, and are limited to voltage phasor estimation, making real-time monitoring of dynamic performance challenging.
Innovation Solution
A method using recursive least square estimation based on real-time PMU and SCADA measurements, independent of grid parameters and state matrix renewal, which finds linear combination relationships between nodes with and without PMU, allowing for real-time estimation of dynamic processes such as voltage, current, and power without requiring complete observability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the Jacobian matrix is recalculated in real-time to estimate dynamic process of nodes without PMU, then the estimation accuracy is improved, but the computing burden becomes too heavy for real-time application
Solution Approach 1:
The Jacobian matrix is pre-calculated offline based on the power flow equations and network topology, and stored for later use. This preliminary computation avoids the need for real-time recalculation, significantly reducing the online computing burden while maintaining estimation accuracy through the use of pre-computed sensitivity relationships.
Solution Approach 2:
The method dynamically updates only the necessary components of the Jacobian matrix based on changing operating conditions, rather than recalculating the entire matrix. This selective update approach adapts to system changes while minimizing computational effort, enabling real-time estimation without full matrix recalculation.
2Productivity
If simplified coefficients are used to reduce computation workload, then the real-time processing capability is improved, but the precision of dynamic process estimation deteriorates
Solution Approach 1:
The method uses dynamic updating of the Jacobian matrix based on actual system operating conditions, allowing the estimation model to adapt to changing states while maintaining computational efficiency. This dynamic approach preserves accuracy by reflecting real system behavior without requiring full recalculation at each time step.
Solution Approach 2:
The estimation method incorporates feedback from actual measurements and system state changes to continuously refine the dynamic process estimates. This feedback mechanism compensates for simplifications in the model, maintaining precision by adjusting estimates based on observed deviations from predicted behavior.
3Ease of manufacture
If rated values and typical parameters of grid devices are used, then the parameter acquisition is simplified, but the accuracy of linear correlation coefficients deteriorates due to parameter inaccuracy
Solution Approach 1:
The method uses feedback from actual system measurements to identify and correct parameter deviations from rated values. By comparing estimated values with actual measurements, the system can detect parameter inaccuracies and adjust the model accordingly, compensating for the use of nominal parameter values without requiring detailed knowledge of actual device parameters.
4Loss of information
If the existing method is used to estimate voltage phasors, then the dynamic process of nodes without PMU can be observed, but the method is limited to voltage phasor estimation and cannot estimate other measurements such as power
Solution Approach 1:
The estimation method is extended to handle multiple types of measurements beyond voltage phasors, including power, frequency, and other system variables. By formulating a unified estimation framework that can accommodate different measurement types, the system achieves multi-functionality and can estimate various dynamic processes simultaneously using the same underlying methodology.
Data Source
AI summary
A state-matrix-independent dynamic process estimation method in real-time for weakly observable measurement nodes without Phasor Measurement Unit(PMU) is only dependent on real-time measurement dynamic data of measurement nodes with PMU and measurement data of Supervisory Control And Data Acquisition (SCADA) system in electric power system or state estimation data. According to the SCADA measurement data or state estimation data at some continuous moments, the method utilizes recursive least squares solution to find a linear combination relationship between variation of measurement parameter to be estimated of nodes without PMU and variation of corresponding measurement parameter of nodes with PMU. Using the linear combination of relationship, the dynamic process of measurement nodes without PMU is estimated in real-time. The method provides high estimation precision and meets error requirements of engineering application.