Power-grid stability prediction using transferable machine learnings

@article{Yang2021PowergridSP,
  title={Power-grid stability prediction using transferable machine learnings},
  author={Seong-Gyu Yang and Beom Jun Kim and Seung-Woo Son and Heetae Kim},
  journal={Chaos},
  year={2021},
  volume={31 12},
  pages={
          123127
        }
}
Complex network analyses have provided clues to improve power-grid stability with the help of numerical models. The high computational cost of numerical simulations, however, has inhibited the approach, especially when it deals with the dynamic properties of power grids such as frequency synchronization. In this study, we investigate machine learning techniques to estimate the stability of power-grid synchronization. We test three different machine learning algorithms-random forest, support… 

Figures and Tables from this paper

Predicting Dynamic Stability of Power Grids using Graph Neural Networks
TLDR
It is shown that SNBS can be predicted in general and the performance significantly changes using different GNN-models, and interesting transfer capabilities of the approach are observed: Gnn-models trained on smaller grids can directly be applied on larger grids without the need of retraining.
Predicting basin stability of power grids using graph neural networks
TLDR
It is shown that SNBS can be predicted in general and the performance significantly changes using different GNN-models, and interesting transfer capabilities of the approach are observed: Gnn-models trained on smaller grids can directly be applied on larger grids without the need of retraining.
Fast Transient Stability Prediction Using Grid-informed Temporal and Topological Embedding Deep Neural Network
Transient stability prediction is critically essential to the fast online assessment and maintaining the stable operation in power systems. The wide deployment of phasor measurement units (PMUs)

References

SHOWING 1-10 OF 76 REFERENCES
Prediction of Power Grid Failure Using Neural Network Learning
  • Carmen Haseltine, E. Eman
  • Physics
    2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)
  • 2017
TLDR
This research study demonstrates that overall “pre-event” analysis can be beneficial with the use of a machine learning agent and can also be used to determine areas that need the most attention.
On structural and dynamical factors determining the integrated basin instability of power-grid nodes.
TLDR
Both structural properties such as gate keepers in network topology and dynamical propertiessuch as large power input/output at nodes cause synchronization instability, and evenly distributed power generation and avoidance of bottlenecks can improve the overall synchronization stability of power-grid systems.
Enhancing power grid synchronization and stability through time-delayed feedback control.
TLDR
An extended version of the Kuramoto model with inertia is considered that includes the voltage dynamics, thus taking into account the interplay of amplitude and phase typical of the electrodynamical behavior of a machine.
Multistability and variations in basin of attraction in power-grid systems
TLDR
It is shown that the observed nonmonotonic behavior is a consequence of transitions in multistability, which are related to changes in stability of the \emph{unsynchronized} states, and suggest that a comprehensive understanding of changes inMultistability are necessary to prevent the unexpected catastrophic instability in the building blocks of power grids.
Abruptness of Cascade Failures in Power Grids
TLDR
It is indicated that increasing the system size causes breakdowns to become more abrupt; in fact, mapping the system to a solvable statistical-physics model indicates the occurrence of a first order transition in the large size limit.
Impact of network topology on synchrony of oscillatory power grids.
TLDR
This work analyzes the impact of grid topologies on spontaneous synchronization, considering regular, random, and small-world topologies and focusing on the influence of decentralization, finding that decentralization itself may support power grids in exhibiting a stable state for lower transmission line capacities.
Community consistency determines the stability transition window of power-grid nodes
TLDR
It is found that nodes with an exceptionally large transition window also have a low community consistency, which gives an efficient way to identify nodes with a long transition window (which is computationally time consuming).
Dynamical modeling of cascading failures in the Turkish power grid.
TLDR
It is found that the network damage depends on the load and generation distribution in the network with centralized generation being more susceptible to failures than a decentralized one and economic considerations on transmission line capacity are shown to conflict with stability.
How dead ends undermine power grid stability.
TLDR
It is found in numerical simulations of artificially generated power grids that tree-like connection schemes--so-called dead ends and dead trees--strongly diminish stability, which may indicate a topological design principle for future power grids: avoid dead ends.
A Machine Learning‐Based Global Atmospheric Forecast Model
TLDR
This paper investigates the applicability of machine learning to weather prediction by building a low-resolution ML model for global weather prediction and the forecast performance of the ML model is compared to conventional models.
...
...