Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Ensuring Data Privacy in AC Optimal Power Flow with a Distributed Co-Simulation Framework

Xinliang Dai, Alexander Kocher, Jovana Kovačević, Burak Dindar, Yuning Jiang, Colin N. Jones, Hüseyin Çakmak, Veit Hagenmeyer

TL;DR

This work addresses data privacy in AC OPF for cross-operator coordination by integrating the convergence-guaranteed ALADIN algorithm with a geographically distributed co-simulation framework (eCosim) inside the eASiMOV platform. It models the problem with region-wise decomposition, core/copy buses, and affine consensus, and demonstrates convergence to a centralized optimum while preserving operator data privacy. A Matlab-based implementation within a co-simulation setting shows that privacy-preserving distributed OPF achieves comparable accuracy to centralized solutions, at a modest overhead due to communication and synchronization. The approach enables practical TSO-DSO coordination in distributed energy systems and points to future work on scaling to larger networks and real-world datasets, along with efficiency improvements in the co-simulation infrastructure.

Abstract

During the energy transition, the significance of collaborative management among institutions is rising, confronting challenges posed by data privacy concerns. Prevailing research on distributed approaches, as an alternative to centralized management, often lacks numerical convergence guarantees or is limited to single-machine numerical simulation. To address this, we present a distributed approach for solving AC Optimal Power Flow (OPF) problems within a geographically distributed environment. This involves integrating the energy system Co-Simulation (eCoSim) module in the eASiMOV framework with the convergence-guaranteed distributed optimization algorithm, i.e., the Augmented Lagrangian based Alternating Direction Inexact Newton method (ALADIN). Comprehensive evaluations across multiple system scenarios reveal a marginal performance slowdown compared to the centralized approach and the distributed approach executed on single machines -- a justified trade-off for enhanced data privacy. This investigation serves as empirical validation of the successful execution of distributed AC OPF within a geographically distributed environment, highlighting potential directions for future research.

Ensuring Data Privacy in AC Optimal Power Flow with a Distributed Co-Simulation Framework

TL;DR

This work addresses data privacy in AC OPF for cross-operator coordination by integrating the convergence-guaranteed ALADIN algorithm with a geographically distributed co-simulation framework (eCosim) inside the eASiMOV platform. It models the problem with region-wise decomposition, core/copy buses, and affine consensus, and demonstrates convergence to a centralized optimum while preserving operator data privacy. A Matlab-based implementation within a co-simulation setting shows that privacy-preserving distributed OPF achieves comparable accuracy to centralized solutions, at a modest overhead due to communication and synchronization. The approach enables practical TSO-DSO coordination in distributed energy systems and points to future work on scaling to larger networks and real-world datasets, along with efficiency improvements in the co-simulation infrastructure.

Abstract

During the energy transition, the significance of collaborative management among institutions is rising, confronting challenges posed by data privacy concerns. Prevailing research on distributed approaches, as an alternative to centralized management, often lacks numerical convergence guarantees or is limited to single-machine numerical simulation. To address this, we present a distributed approach for solving AC Optimal Power Flow (OPF) problems within a geographically distributed environment. This involves integrating the energy system Co-Simulation (eCoSim) module in the eASiMOV framework with the convergence-guaranteed distributed optimization algorithm, i.e., the Augmented Lagrangian based Alternating Direction Inexact Newton method (ALADIN). Comprehensive evaluations across multiple system scenarios reveal a marginal performance slowdown compared to the centralized approach and the distributed approach executed on single machines -- a justified trade-off for enhanced data privacy. This investigation serves as empirical validation of the successful execution of distributed AC OPF within a geographically distributed environment, highlighting potential directions for future research.
Paper Structure (13 sections, 7 equations, 5 figures, 1 table, 3 algorithms)

This paper contains 13 sections, 7 equations, 5 figures, 1 table, 3 algorithms.

Table of Contents

  1. Introduction
  2. Distributed AC Optimal Power Flow
  3. Conventional Formulation
  4. Distributed Reformulation
  5. Distributed Optimization Algorithm
  6. Distributed AC OPF with Co-Simulation
  7. Co-Simulation Framework Description
  8. Adaptation to Distributed AC OPF
  9. Implementation Details
  10. Case Study
  11. Distributed AC OPF Co-Simulation Setting
  12. Results and Discussions
  13. Conclusion and Outlook

Figures (5)

  • Figure 1: Decomposition by sharing components between neighboring regions.
  • Figure 2: The easimov-ecosim co-simulation architecture enables geographically distributed AC OPF calculation with respect to data and model privacy.
  • Figure 3: Integration of matlab OPF code (Algorithm 1) into the ecosim control code (Algorithm 2 and 3).
  • Figure 4: Runtime comparison for the use cases with a serial matlab implementation in (a) IPOPT and (b) ALADIN, a distributed execution with ecosim on one computer in the KIT network in (c) eCoSim-KIT1, on five computers in the KIT network in (d) eCoSim-KIT5 and a geographically distributed co-simulation with access to the network storage located at KIT over a VPN connection in (e) eCoSim-Geo5. The clients are distributed over three cities with a geographical distance of up to 15 km to KIT (the internet routing Runtimes are measured at the coordinator software module located at KIT.
  • Figure 5: Numerical Results by proposed distributed algorithm.

Theorems & Definitions (2)

  • Remark 1
  • Remark 2