Active Bucketized Learning for ACOPF Optimization Proxies

@article{Klamkin2022ActiveBL,
  title={Active Bucketized Learning for ACOPF Optimization Proxies},
  author={Michael Klamkin and Mathieu Tanneau and Terrence W.K. Mak and Pascal Van Hentenryck},
  journal={ArXiv},
  year={2022},
  volume={abs/2208.07497}
}
This paper considers optimization proxies for Optimal Power Flow (OPF), i.e., machine-learning models that approximate the input/output relationship of OPF. Recent work has fo- cused on showing that such proxies can be of high fidelity. However, their training requires significant data, each in- stance necessitating the (offline) solving of an OPF for a sample of the input distribution. To meet the requirements of market-clearing applications, this paper proposes Active Bucketized Sampling (ABS… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 45 REFERENCES

Supervised learning for optimal power flow as a real-time proxy

This work designs and compares different supervised learning algorithms to compute the cost of Alternating Current Optimal Power Flow and describes this approximation module as a proxy for predicting short-term decision outcomes without the need of actual simulation and optimization of them.

Learning Optimization Proxies for Large-Scale Security-Constrained Economic Dispatch

Predicting AC Optimal Power Flows: Combining Deep Learning and Lagrangian Dual Methods

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.

Optimal Power Flow Using Graph Neural Networks

Optimal power flow (OPF) is one of the most important optimization problems in the energy industry. In its simplest form, OPF attempts to find the optimal power that the generators within the grid

Learning Optimal Solutions for Extremely Fast AC Optimal Power Flow

  • Ahmed S. ZamzamK. Baker
  • Computer Science
    2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)
  • 2020
A machine learning approach to optimize the real-time operation of electric power grids finds feasible solutions to the AC optimal power flow (OPF) problem with negligible optimality gaps, resulting in a significant decrease in computational burden for grid operators.

BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning

We develop BatchBALD, a tractable approximation to the mutual information between a batch of points and model parameters, which we use as an acquisition function to select multiple informative points

DeepOPF: Deep Neural Network for DC Optimal Power Flow

  • Xiang PanTianyu ZhaoMinghua Chen
  • Computer Science
    2019 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm)
  • 2019
Simulation results of IEEE test cases show that DeepOPF always generates feasible solutions with negligible optimality loss, while speeding up the computing time by two orders of magnitude as compared to conventional approaches implemented in a state-of-the-art solver.

Machine Learning for AC Optimal Power Flow

This work presents two formulations of ACOPF as a machine learning problem: 1) an end-to-end prediction task where the optimal generator settings are predicted, and 2) a constraint predictiontask where the set of active constraints in the optimal solution are predicted.

Warm-Starting AC Optimal Power Flow with Graph Neural Networks

On a synthetic power grid modelling Texas, it is shown that a Graph Neural Networks (GNNs) produces a solution four orders of magnitude faster or can warm-start the optimizer for 3.75x faster convergence.