Predicting AC Optimal Power Flows: Combining Deep Learning and Lagrangian Dual Methods
@inproceedings{Fioretto2019PredictingAO, title={Predicting AC Optimal Power Flows: Combining Deep Learning and Lagrangian Dual Methods}, author={Ferdinando Fioretto and Terrence W.K. Mak and Pascal Van Hentenryck}, booktitle={AAAI Conference on Artificial Intelligence}, year={2019} }
The Optimal Power Flow (OPF) problem is a fundamental building block for the optimization of electrical power systems. It is nonlinear and nonconvex and computes the generator setpoints for power and voltage, given a set of load demands. It is often needed to be solved repeatedly under various conditions, either in real-time or in large-scale studies. This need is further exacerbated by the increasing stochasticity of power systems due to renewable energy sources in front and behind the meter…
Figures and Tables from this paper
85 Citations
High-Fidelity Machine Learning Approximations of Large-Scale Optimal Power Flow
- Engineering, Computer ScienceArXiv
- 2020
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.
Unsupervised Deep Learning for AC Optimal Power Flow via Lagrangian Duality
- Computer ScienceGLOBECOM 2022 - 2022 IEEE Global Communications Conference
- 2022
A deep neural network is developed to output a partial set of decision variables while the remaining variables are recovered by solving AC power flow equations and the fast decoupled power flow solver is adopted to further reduce the computational time.
A learning-augmented approach for AC optimal power flow
- Computer Science, Engineering
- 2021
Combining Deep Learning and Optimization for Preventive Security-Constrained DC Optimal Power Flow
- Computer ScienceIEEE Transactions on Power Systems
- 2021
A novel approach that combines deep learning and robust optimization techniques is proposed that predicts directly the SCOPF implementable solution and results in significant time reduction for obtaining feasible solutions with an optimality gap below 0.1%.
Combining Deep Learning and Optimization for Security-Constrained Optimal Power Flow
- Computer ScienceArXiv
- 2020
A novel approach that combines deep learning and robust optimization techniques is proposed that predicts directly the SCOPF implementable solution and results in significant time reduction for obtaining feasible solutions with an optimality gap below 0.1%.
Learning to Solve the AC-OPF Using Sensitivity-Informed Deep Neural Networks
- Computer ScienceIEEE Transactions on Power Systems
- 2022
Numerical tests on three benchmark power systems corroborate the advanced generalization and constraint satisfaction capabilities for the OPF solutions predicted by an SI-DNN over a conventionally trained DNN, especially in low-data setups.
Learning to Optimize Power Distribution Grids using Sensitivity-Informed Deep Neural Networks
- Computer Science2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)
- 2020
Numerical tests showcase that sensitivity-informed deep learning can enhance prediction accuracy in terms of mean square error (MSE) by 2-3 orders of magnitude at minimal computational overhead.
DeepOPF: A Feasibility-Optimized Deep Neural Network Approach for AC Optimal Power Flow Problems
- Computer ScienceIEEE Systems Journal
- 2022
An efficient Deep Neural Network approach, DeepOPF, is developed to ensure the feasibility of the generated solution of the AC-OPF problem, by employing a penalty approach in training the DNN.
Fast DC Optimal Power Flow Based on Deep Convolutional Neural Network
- Engineering2022 IEEE 5th International Electrical and Energy Conference (CIEEC)
- 2022
The optimal power flow is the cornerstone of the operation and management of electric power systems. However, the stochastic and intermittent uncertainty due to the proliferation of renewable energy…
Learning to Solve AC Optimal Power Flow by Differentiating through Holomorphic Embeddings
- EngineeringArXiv
- 2020
This approach constitutes the first learning-based approach that successfully respects the full non-linear AC-OPF equations and reports a 12x increase in speed and a 40% increase in robustness compared to a traditional solver.
References
SHOWING 1-10 OF 30 REFERENCES
Learning for DC-OPF: Classifying active sets using neural nets
- Computer Science2019 IEEE Milan PowerTech
- 2019
This paper proposes the use of classification algorithms to learn the mapping between the uncertainty realization and the active set of constraints at optimality, thus further enhancing the computational efficiency of the real-time prediction of the optimal power flow.
Statistical Learning for DC Optimal Power Flow
- Computer Science2018 Power Systems Computation Conference (PSCC)
- 2018
An ensemble control policy is proposed that combines several basis policies to improve performance and is based on the observation that the OPF solution corresponding to a certain uncertainty realization is a basic feasible solution, which provides an affine control policy.
NESTA, The NICTA Energy System Test Case Archive
- EngineeringArXiv
- 2014
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.
PowerModels.J1: An Open-Source Framework for Exploring Power Flow Formulations
- Computer Science2018 Power Systems Computation Conference (PSCC)
- 2018
This work proposes PowerModels, an open-source platform for comparing power flow formulations, and provides a brief introduction to the design, validates its implementation, and demonstrates its effectiveness with a proof-of-concept study analyzing five different formulations of the Optimal Power Flow problem.
Real-Time Optimal Power Flow
- Engineering, Computer ScienceIEEE Transactions on Smart Grid
- 2017
This paper builds on recent work to develop a real-time algorithm for AC optimal power flow, based on quasi-Newton methods, that uses second-order information to provide suboptimal solutions on a fast timescale, and can be shown to track the optimalPower flow solution when the estimated second- order information is sufficiently accurate.
Enhancement of hybrid renewable energy systems control with neural networks applied to weather forecasting: the case of Olvio
- EngineeringNeural Computing and Applications
- 2015
The RNN framework, trained with local meteorological data, successfully manages to enhance and optimize the PMS based on the provided solar radiation and wind speed prediction and make the specific HYRES suitable for use as a stand-alone remote energy plant.
LEAP nets for power grid perturbations
- Computer ScienceESANN
- 2019
We propose a novel neural network embedding approach to model power transmission grids, in which high voltage lines are disconnected and reconnected with one-another from time to time, either…
Transmission expansion planning: A review
- Engineering2016 International Conference on Energy Efficient Technologies for Sustainability (ICEETS)
- 2016
Modern electric power systems consist of large-scale highly complex interconnected transmission systems and, thus, transmission expansion planning (TEP) has become a significant power system…
Deep Learning for Power System Security Assessment
- Computer Science2019 IEEE Milan PowerTech
- 2019
This paper represents for the first time power system snapshots as 2-dimensional images, thus taking advantage of the wide range of deep learning methods available for image processing, and finds that its approach is over 255 times faster than a standard small-signal stability assessment.
A Coordinate-Descent Algorithm for Tracking Solutions in Time-Varying Optimal Power Flows
- Computer Science2018 Power Systems Computation Conference (PSCC)
- 2018
This work proposes to use a coordinate-descent algorithm for solving time-varying optimisation problems of transmission-constrained problems in power systems, and bound the difference between the current approximate optimal cost generated by the algorithm and the optimal cost for a relaxation using the most recent data from above.