IGNNITION: Bridging the Gap between Graph Neural Networks and Networking Systems

Pujol-Perich, David; Suárez-Varela, José; Ferriol, Miquel; Xiao, Shihan; Cabellos‐Aparicio, Albert; Barlet‐Ros, Pere

doi:10.1109/mnet.001.2100266

Cited by 19 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GNN models are tested in situations in which the number of PMUs is optimal (with minimal measurement redundancy under which the WLS SE provides a solution), in underdetermined scenarios, as well as in scenarios with maximal measurement redundancy. We used the IGNNITION framework [25] for…”

Section: Numerical Resultsmentioning

confidence: 99%

Robust and Fast Data-Driven Power System State Estimator Using Graph Neural Networks

Ognjen¹,

Cosovic²,

Vukobratović³

2022

Preprint

View full text Add to dashboard Cite

The power system state estimation (SE) algorithm estimates the complex bus voltages based on the available set of measurements. Because phasor measurement units (PMUs) are becoming more widely employed in transmission power systems, a fast SE solver capable of exploiting PMUs' high sample rates is required. To accomplish this, we present a method for training a model based on graph neural networks (GNNs) to learn estimates from PMU voltage and current measurements, which, once it is trained, has a linear computational complexity with respect to the number of nodes in the power system. We propose an original GNN implementation over the power system's factor graph to simplify the incorporation of various types and numbers of measurements both on power system buses and branches. Furthermore, we augment the factor graph to improve the robustness of GNN predictions. Training and test examples were generated by randomly sampling sets of power system measurements and annotated with the exact solutions of linear SE with PMUs. The numerical results demonstrate that the GNN model provides an accurate approximation of the SE solutions. Additionally, errors caused by PMU malfunctions or the communication failures that make the SE problem unobservable have a local effect and do not deteriorate the results in the rest of the power system.

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

Robust and Fast Data-Driven Power System State Estimator Using Graph Neural Networks

Ognjen¹,

Cosovic²,

Vukobratović³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In this section, we describe the GNN model's training process and test the trained model on various examples to validate its accuracy, and its robustness under measurement data loss due to communication failure and cyber attacks in the form of malicious data injections. The described GNN model for augmented factor graphs is implemented using the IGNNITION library [22] and trained for 100 epochs in all experiments using the following hyperparameters: 64 elements in the node embedding vector, 32 graphs in a mini-batch, four GNN layers, ReLU activation functions, Adam optimiser, and a learning rate of 4 × 10 −4 . We conducted separate training experiments for IEEE 30 and IEEE 118-bus test cases, for which we generated a training set containing 10000 samples and validation and test sets containing 100 samples each.…”

Section: Numerical Results and Discussionmentioning

confidence: 99%

Distributed Nonlinear State Estimation in Electric Power Systems using Graph Neural Networks

Ognjen¹,

Cosovic²,

Miskovic³

et al. 2022

Preprint

View full text Add to dashboard Cite

Nonlinear state estimation (SE), with the goal of estimating complex bus voltages based on all types of measurements available in the power system, is usually solved using the iterative Gauss-Newton method. The nonlinear SE presents some difficulties when considering inputs from both phasor measurement units and supervisory control and data acquisition system. These include numerical instabilities, convergence time depending on the starting point of the iterative method, and the quadratic computational complexity of a single iteration regarding the number of state variables. This paper introduces an original graph neural network based SE implementation over the augmented factor graph of the nonlinear power system SE, capable of incorporating measurements on both branches and buses, as well as both phasor and legacy measurements. The proposed regression model has linear computational complexity during the inference time once trained, with a possibility of distributed implementation. Since the method is noniterative and non-matrix-based, it is resilient to the problems that the Gauss-Newton solver is prone to. Aside from prediction accuracy on the test set, the proposed model demonstrates robustness when simulating cyber attacks and unobservable scenarios due to communication irregularities. In those cases, prediction errors are sustained locally, with no effect on the rest of the power system's results.

show abstract

“…In this section, we conduct comprehensive numerical tests to evaluate the effectiveness of proposed augmented factor graph-based GNN approaches for linear and nonlinear SE problems. We used the IGNNITION framework [151] for building and utilizing GNN models, with the hyperparameters presented in Table 4.1, the first three of which were obtained with the grid search hyperparameter optimization using the Tune tool [152]. All the results presented in this section are normalized using the corresponding nominal voltages in the test power systems and a base power of 100 MVA.…”

Section: Numerical Resultsmentioning

confidence: 99%

Near Real-Time Distributed State Estimation via AI/ML-Empowered 5G Networks

Ognjen¹,

Forcan

Cosovic

et al. 2022

2022 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)

View full text Add to dashboard Cite

Electric power systems consist of generation, distribution, and transmission systems, which are all traditionally coordinated from the corresponding control centres. System operators use specialized software solutions for monitoring and optimization of electric power systems, installed in control centres. Typical algorithms implemented in mentioned software solutions should satisfy near real-time operation requirements, while delivering accurate information for power system monitoring and optimizing its operation.Modern electric power systems have been increasing in size, complexity, as well as dynamics due to the growing integration of renewable energy resources, which have sporadic power generation. This necessitates the development of near real-time power system algorithms, demanding lower computational complexity regarding the power system size. Considering the growing trend in the collection of historical measurement data and recent advances in the rapidly developing deep learning field, the topic of this dissertation is the application of deep learning algorithms, namely graph neural networks (GNNs) and deep reinforcement learning (DRL), for monitoring and optimization of electric power systems.

show abstract

IGNNITION: Bridging the Gap between Graph Neural Networks and Networking Systems

Cited by 19 publications

References 7 publications

Robust and Fast Data-Driven Power System State Estimator Using Graph Neural Networks

Robust and Fast Data-Driven Power System State Estimator Using Graph Neural Networks

Distributed Nonlinear State Estimation in Electric Power Systems using Graph Neural Networks

Near Real-Time Distributed State Estimation via AI/ML-Empowered 5G Networks

Contact Info

Product

Resources

About