Method for assessing short-term voltage stability of power grid based on synchronous spatio-temporal information learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for assessing short-term voltage stability in power grids lack an effective synchronous processing mechanism, leading to information loss and reduced reliability in stability assessment models, particularly in densely populated urban areas where transient faults can cause voltage instability and collapse.
Innovation Solution
A method using synchronous spatio-temporal information learning, involving time-domain simulations, construction of spatial and temporal adjacency matrices, and graph convolutional neural networks to build a short-term voltage stability assessment model that synchronizes spatio-temporal information for real-time online monitoring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If asynchronous discrete learning methods are used for spatio-temporal information, then the complexity of the learning process is reduced, but information loss occurs and assessment reliability deteriorates
Solution Approach 1:
The patent merges spatial and temporal information processing into a unified synchronous learning framework. The spatio-temporal graph convolutional neural network integrates spatial adjacency matrices (representing grid connectivity) and temporal adjacency matrices (representing time-series correlations) into a single model structure, enabling simultaneous processing of both dimensions without information loss while maintaining manageable complexity through modular architecture design.
Solution Approach 2:
The patent transforms the learning approach from separate discrete processing of spatial and temporal dimensions to a unified spatio-temporal dimension. By constructing a spatio-temporal adjacency matrix that combines spatial relationships (grid topology) and temporal relationships (time evolution), the model operates in an enhanced dimensional space that captures comprehensive correlations without losing information from either dimension.
2Reliability
If synchronous spatio-temporal information learning is implemented, then information loss is prevented and assessment reliability is improved, but the complexity of the learning mechanism increases
Solution Approach 1:
The patent segments the complex spatio-temporal learning mechanism into distinct functional modules: (1) spatial adjacency matrix construction based on grid topology, (2) temporal adjacency matrix construction based on time-series correlations, (3) spatio-temporal graph convolutional neural network processing, and (4) stability assessment output. This segmentation allows each module to be optimized independently while maintaining their synergistic integration, thereby managing overall complexity.
Solution Approach 2:
The spatio-temporal graph convolutional neural network serves multiple functions simultaneously: it processes spatial relationships between grid nodes, captures temporal evolution of system states, performs feature extraction from multi-dimensional data, and generates stability assessments. This multi-functionality reduces the need for separate specialized components, thereby managing complexity while achieving comprehensive analysis.
3Measurement precision
If comprehensive spatio-temporal correlation characterization is performed, then the accuracy of stability prediction is improved, but the computational requirements and processing complexity increase
Solution Approach 1:
The patent performs preliminary construction of spatial and temporal adjacency matrices before the main learning process. The spatial adjacency matrix is pre-computed based on the known grid topology, and the temporal adjacency matrix is pre-computed based on historical time-series data. This preliminary action organizes the data structure in advance, reducing the computational burden during real-time processing while maintaining comprehensive spatio-temporal correlation characterization for high prediction accuracy.
Data Source
AI summary
A method for assessing short-term voltage stability of a power grid based on synchronous spatio-temporal information learning includes: performing a time-domain simulation for each transient fault of a power grid under each operating mode, and extracting temporal response trajectories of each monitoring bus and stability status information of the power grid to obtain a transient sample set; constructing a spatial adjacency matrix and a temporal adjacency matrix separately, and integrating the spatial adjacency matrix and the temporal adjacency matrix into a spatio-temporal adjacency matrix; performing synchronous spatio-temporal information learning by using a graph convolutional neural network algorithm, and performing training to obtain a short-term voltage stability assessment model driven by spatio-temporal information synchronization; and inputting a transient temporal responsive trajectory obtained into the short-term voltage stability assessment model, to obtain an assessment result of short-term voltage stability of the power grid.