Deep Learning for Power System Security Assessment

@article{HidalgoArteaga2019DeepLF,
  title={Deep Learning for Power System Security Assessment},
  author={Jos{\'e}-Mar{\'i}a Hidalgo-Arteaga and Fiodar Hancharou and Florian Thams and Spyros Chatzivasileiadis},
  journal={2019 IEEE Milan PowerTech},
  year={2019},
  pages={1-6}
}
Security assessment is among the most fundamental functions of power system operator. The sheer complexity of power systems exceeding a few buses, however, makes it an extremely computationally demanding task. The emergence of deep learning methods that are able to handle immense amounts of data, and infer valuable information appears as a promising alternative. This paper has two main contributions. First, inspired by the remarkable performance of convolutional neural networks for image… 

Figures and Tables from this paper

Verification of Neural Network Behaviour: Formal Guarantees for Power System Applications
TLDR
Developing a rigorous framework based on mixed-integer linear programming, these methods can determine the range of inputs that neural networks classify as safe or unsafe, and are able to systematically identify adversarial examples.
Verication of Neural Network Behaviour: Formal Guarantees for Power System Applications
TLDR
Developing a rigorous framework based on mixed-integer linear programming, these methods can determine the range of inputs that neural networks classify as safe or unsafe, and are able to systematically identify adversarial examples.
DeVLearn: A Deep Visual Learning Framework for Determining the Location of Temporary Faults in Power Systems
TLDR
The results show that for faults on two distinct lines in the IEEE 68-bus network, DeVLearn is able to project PMU measurements into a two-dimensional space such that data for faults at different locations separate into well-defined clusters.
DeVLearn: A Deep Visual Learning Framework for Localizing Temporary Faults in Power Systems
TLDR
The results show that for faults on two different lines in the IEEE 68-bus network, DeVLearn is able to project PMU measurements into a two-dimensional space such that data for faults at different locations separate into well-defined clusters.
Predicting AC Optimal Power Flows: Combining Deep Learning and Lagrangian Dual Methods
TLDR
A deep learning approach to the Optimal Power Flow problem that exploits the information available in the prior states of the system, as well as a dual Lagrangian method to satisfy the physical and engineering constraints present in the OPF.
Efficient Creation of Datasets for Data-Driven Power System Applications
High-Fidelity Machine Learning Approximations of Large-Scale Optimal Power Flow
TLDR
This paper proposes an integration of deep neural networks and Lagrangian duality to capture the physical and operational constraints of the AC Optimal Power Flow and produces highly accurate approximations whose costs are within 0.01% of optimality.
Versatile and Robust Transient Stability Assessment via Instance Transfer Learning
TLDR
A new data collection method in a data-driven algorithm incorporating the knowledge of power system dynamics called Fault-Affected Area, which provides crucial information regarding the unstable region of operation to support N-1 pre-fault transient stability assessment.
Physics-Informed Neural Networks for Power Systems
TLDR
This work unlocks a range of opportunities in power systems, being able to determine dynamic states, such as rotor angles and frequency, and uncertain parameters such as inertia and damping at a fraction of the computational time required by conventional methods.
Guiding Cascading Failure Search with Interpretable Graph Convolutional Network
TLDR
This work shows that the complex mechanism of cascading failures can be well captured by training a graph convolutional network (GCN) offline, and can be significantly accelerated with the aid of the trained GCN model.
...
...

References

SHOWING 1-10 OF 20 REFERENCES
A Deep Learning-Based Feature Extraction Framework for System Security Assessment
TLDR
It is shown how deep autoencoders can be used to transform the space of conventional state variables to a small number of dimensions where the authors can optimally distinguish between safe and unsafe operation.
Efficient Database Generation for Data-Driven Security Assessment of Power Systems
TLDR
A modular and highly scalable algorithm for computationally efficient database generation using convex relaxation techniques and complex network theory to discard large infeasible regions and drastically reduce the search space.
Power System Security Assessment Using Neural Networks: Feature Selection Using Fisher Discrimination
TLDR
This paper investigates the use of Fisher's linear discriminant function, coupled with feature selection techniques as a means for selecting neural network training features for power system security assessment.
NESTA, The NICTA Energy System Test Case Archive
TLDR
This report surveys all of the publicly available AC transmission system test cases, to the best of the authors' knowledge, and finds that many of the traditional test cases are missing key network operation constraints, such as line thermal limits and generator capability curves.
Power system security assessment
Security refers to the degree of risk in a power system's ability to survive imminent disturbances (contingencies) without interruption to customer service. It relates to robustness of the system to
Automatic Learning Techniques in Power Systems
TLDR
This book presents a representative subset of automatic learning methods - basic and more sophisticated ones - available from statistics, and from artificial intelligence, and appropriate methodologies for combining these methods to make the best use of available data in the context of real-life problems.
Data-Driven Security-Constrained OPF
In this paper we unify electricity market operations with power system security considerations. Using data-driven techniques, we address both small signal stability and steady-state security, derive
Gradient-based learning applied to document recognition
TLDR
This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task, and Convolutional neural networks are shown to outperform all other techniques.
Deep Learning using Rectified Linear Units (ReLU)
TLDR
The use of rectified linear units (ReLU) as the classification function in a deep neural network (DNN) is introduced by taking the activation of the penultimate layer of a neural network, then multiplying it by weight parameters $\theta$ to get the raw scores.
Adam: A Method for Stochastic Optimization
TLDR
This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
...
...