Quantum mechanics

Algorithm

Residual

Control (management)

Scheme (mathematics)

Voltage

Engineering

Curvature

Convergence (economics)

Distributed algorithm

Computation

Economics

Power (physics)

Scalability

Computer science

Mathematical optimization

Distributed generation

Distributed computing

Regression analysis

Physics

AC power

Database

Power flow

Mathematics

Test case

Distributed power

Flow (mathematics)

Electric power system

Power-flow study

Mathematical analysis

Artificial intelligence

Geometry

Economic growth

Control theory (sociology)

Mechanics

Economic dispatch

Machine learning

Electrical engineering

IEEE Transactions on Power Systems

In light of the soaring uptake of distributed energy resources, distributed methods are attracting increased focus. This paper proposes an adaptive scheme to improve the convergence of the alternating direction of multipliers method (ADMM) on the component-based dual decomposition of the AC optimal power flow (OPF). The method incorporates a local curvature approximation scheme, with underlying parameter update steps inspired by the local residual balancing scheme, to automatically tune the penalty parameters locally, without central oversight. Substantial speed-ups over the vanilla ADMM, as high as 10.6 times, are demonstrated on real-world test systems with more than 13 600 buses, as well as on other difficult test cases. Furthermore, scalability is assessed by showing that, unlike in the centralized computation, the distributed implementation of the proposed adaptive method is independent of system size.

Adaptive ADMM for Distributed AC Optimal Power Flow