Transient Stability Assessment using Gated Recurrent Unit
Authors:
- 1 June 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 560-564
Abstract
Machine learning based transient stability analysis of power system mostly uses feedforward neural network, which does not consider the relationship between data and data in time domain. This paper presents a power system transient stability assessment based on gated recurrent unit. The proposed system manages to extract both spatial and temporal data dependency from the input power system state for security assessment, and output evaluation value faster by controlling stability threshold η.In this paper, the active power and reactive power of the line are input, which are obtained by time-domain simulation. Two different evaluation methods are used to evaluate the experimental results. The simulation results on IEEE 39-bus test system demonstrate accurate assessment with significantly reduced system response time.Keywords
This publication has 8 references indexed in Scilit:
- Research on Text Classification Based on CNN and LSTM2019
- Fast Transient Stability Batch Assessment Using Cascaded Convolutional Neural NetworksIEEE Transactions on Power Systems, 2019
- Intelligent Time-Adaptive Transient Stability Assessment SystemIEEE Transactions on Power Systems, 2017
- Power System Transient Stability Assessment Based on Big Data and the Core Vector MachineIEEE Transactions on Smart Grid, 2016
- Robust Online Dynamic Security Assessment Using Adaptive Ensemble Decision-Tree LearningIEEE Transactions on Power Systems, 2013
- Support vector machine-based algorithm for post-fault transient stability status prediction using synchronized measurements2011
- Transient Stability of Power Systems2000
- A simple direct method for fast transient stability assessment of large power systemsIEEE Transactions on Power Systems, 1988