⚡ electrical engineering

Fair Dynamic Operating Envelopes using Distributed Multi-Period Optimal Power Flow and Jain Index for Active Distribution Networks

This paper proposes a two-stage, multi-period framework that combines distributed optimal power flow with a Jain index-based fairness mechanism to generate dynamic operating envelopes for active distribution networks, effectively balancing network feasibility, equity, and efficiency while managing renewable curtailment and uncertainty through battery storage.

Original authors: Pedro Salomão Quessongo, Daniel Gebbran, Clodomiro Unsihuay-Vila

Published 2026-08-25

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Pedro Salomão Quessongo, Daniel Gebbran, Clodomiro Unsihuay-Vila

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The electricity grid that powers our homes and businesses is undergoing a quiet revolution. For decades, these networks were designed as one-way streets, sending power from large, centralized power plants down to passive consumers. Today, however, millions of households and businesses are becoming "prosumers," generating their own electricity with rooftop solar panels and wind turbines, and storing it in batteries. This shift turns the grid into a complex, two-way system where power flows in many directions at once. While this is a triumph for renewable energy, it creates a new challenge: too much power in the wrong place at the wrong time can overload the wires and damage the system. To manage this, grid operators use "dynamic operating envelopes," which are essentially time-varying limits on how much electricity a prosumer can send back to the grid. If a neighborhood is already receiving too much solar power, the envelope tightens, telling some generators to hold back. The problem is that these limits have traditionally been set purely on technical grounds, which often means that people living at the end of a long power line get stricter limits than those closer to the substation, simply because of their location. This creates an unfair situation where some neighbors are forced to waste their clean energy while others are allowed to sell it all, raising questions about equity in the energy transition.

Researchers at the Federal University of Paraná in Brazil have developed a new method to solve this problem, creating a system that balances the physical safety of the grid with a genuine sense of fairness. Their approach, tested on a simulated version of a standard electrical network, separates the technical limits of the grid from the social goal of fairness. First, they calculate the maximum amount of power the network can physically handle without breaking, a purely technical ceiling. Then, they apply a second layer of logic that redistributes this capacity. Instead of letting the grid's physics decide who gets cut off, their system looks at the entire day's history of energy production and curtailment. It ensures that no single prosumer is unfairly penalized over time by capping the total amount of energy any one person is forced to waste. This is achieved through a two-stage process: first, the system determines the safe technical limits, and second, it adjusts the individual limits so that the burden of reducing energy is shared more equally among all participants, even if it means slightly more total energy is wasted overall to achieve that balance.

The researchers tested this framework on a model of a 33-bus radial feeder, a standard test network used to simulate a typical local distribution system, over a full 24-hour cycle. In their simulations, they introduced various scenarios, including days with high solar output, high wind, and fluctuating demand, while also accounting for the presence of battery storage systems that can shift energy from sunny afternoons to cloudy evenings. The results showed a clear trade-off. When the researchers used only the traditional technical method, the system curtailed, or wasted, about 2.11 megawatt-hours of renewable energy over the day. However, when they applied their new fairness rules, the total amount of wasted energy increased to 5.72 megawatt-hours. This might seem like a negative outcome at first glance, but the nature of that waste changed dramatically. Under the old method, the waste was unevenly distributed, with some prosumers losing nearly all their potential energy while others lost very little. Under the new method, the waste was spread out much more evenly. The researchers measured this using a standard metric for equality, known as the Jain fairness index, which reached values very close to one, indicating near-perfect equality in how the burden was shared. The maximum cumulative amount of energy any single prosumer was forced to give up was capped at just 11.20 percent of their available generation.

Crucially, the study demonstrated that this fairness could be achieved without breaking the grid. The researchers used a sophisticated mathematical technique to solve the problem across different parts of the network simultaneously, ensuring that the proposed limits were physically possible. They then double-checked their results using a more complex, realistic model of electricity flow to confirm that voltages remained stable and no wires overheated. The simulations showed that even with the stricter, fairer limits, the voltage levels stayed within safe boundaries, and no thermal violations occurred. The system also successfully utilized battery storage to help manage the flow, charging up during the day when solar power was abundant and discharging in the evening when demand peaked. This storage acted as a buffer, allowing the system to absorb some of the extra energy that would otherwise have to be curtailed. The study found that while the fairness constraints did increase the total amount of energy that had to be left unused, the battery systems helped mitigate the impact, proving that fairness and technical reliability can coexist if the system is designed with both in mind from the start.

The significance of this work lies in its ability to make the cost of fairness visible and manageable. In previous approaches, fairness was often hidden inside a complex calculation, making it difficult to see exactly how much extra energy had to be sacrificed to ensure everyone was treated equally. By separating the technical limits from the fairness rules, the researchers created a transparent framework where grid operators can see exactly how much efficiency is being traded for equity. They found that by setting a clear budget for how much extra curtailment is acceptable, they could redistribute capacity in a way that prevented the systematic discrimination of prosumers located at the far ends of the network. The study suggests that for modern grids to be truly sustainable, they must be designed not just to be efficient, but also to be fair. The results indicate that a multi-period approach, which considers the entire day's history rather than just a single moment, is essential for managing these complex systems. Without this long-term view, batteries and fairness rules cannot work together effectively, and the grid risks becoming either unsafe or deeply unjust. This research provides a concrete path forward, showing that it is possible to build an energy system where the lights stay on, the wires stay safe, and every neighbor gets a fair shot at using their own clean energy.

Technical Summary: Fair Dynamic Operating Envelopes using Distributed Multi-Period Optimal Power Flow and Jain Index

Problem Statement
The increasing penetration of Distributed Energy Resources (DERs), such as photovoltaic (PV) generation, wind, and battery energy storage systems (BESS), transforms passive distribution networks into active systems with bidirectional power flows. While Dynamic Operating Envelopes (DOEs) are used to define time-varying export limits to maintain network integrity, purely technical DOE allocation often systematically privileges prosumers at electrically favorable locations, leading to uneven renewable curtailment. Furthermore, embedding fairness directly into a single-period Optimal Power Flow (OPF) objective conflates network feasibility, equity, and efficiency, obscuring the specific cost of fairness. Additionally, existing multi-period formulations often fail to separate network-feasible technical envelopes from explicit fairness constraints and storage recourse, particularly when storage couples successive operating intervals through the state of charge (SOC).

Methodology
The paper proposes a two-stage, multi-period framework designed to decouple technical feasibility from equity considerations while accounting for uncertainty and storage dynamics.

Two-Stage DOE Computation:
- Stage 1 (Technical): A distributed multi-period OPF (DOPF) based on a lossless LinDistFlow model computes network-feasible technical export envelopes ( $d^{tech}_{i,t}$ ) for all prosumers, respecting voltage and thermal limits.
- Stage 2 (Fairness): A dynamic aggregate export budget ( $D^{cap}_t = \beta_t D^{tech}_t$ ) is applied, where $\beta_t \in [0,1]$ can restrict capacity below the technical maximum. A fairness subproblem redistributes capacity using a cumulative proportional fairness criterion. This stage minimizes the maximum cumulative curtailment ratio ( $\Gamma$ ) across prosumers, subject to an admissible efficiency budget that limits the total increase in curtailment relative to the technical benchmark.
Operational Scheme with Recourse:
- The resulting fair DOEs ( $d^{fair}_{i,t}$ ) are treated as first-stage, scenario-independent decisions.
- BESS units act as recourse variables, adapting their charging/discharging schedules ( $P^{ch}_{b,t}, P^{dis}_{b,t}$ ) and SOC to specific demand and renewable uncertainty scenarios ( $\omega$ ) to maintain feasibility.
Distributed Solution and Validation:
- The operational problem is solved using a calibrated regional Alternating Direction Method of Multipliers (ADMM). The feeder is partitioned into regions, with consensus variables (active power, reactive power, and squared voltages) exchanged at interfaces.
- Crucially, the linearized LinDistFlow dispatch is independently validated using a nonlinear AC power flow solver to verify physical feasibility (voltage magnitudes and thermal limits) without re-optimizing.
Fairness Metrics:
- Fairness is quantified using the Jain Fairness Index (JFI) on renewable acceptance ratios and the Gini coefficient on cumulative curtailment energy.

Key Contributions

Decoupled Framework: The paper introduces a workflow that separates network-feasible technical envelopes from an explicit, budgeted fairness stage. This allows for the quantification of the equity-efficiency trade-off under storage coupling and distributed coordination.
Multi-Period Fairness with Storage: It proposes a multi-period operational scheme where fair DOEs are fixed first-stage decisions, while BESS provides scenario-dependent recourse. This addresses the issue that single-interval fairness definitions can accumulate uneven curtailment over time.
Calibrated Distributed Solution: The study implements a regional ADMM solution on a partitioned radial feeder using a LinDistFlow model, coupled with independent nonlinear AC validation for every operating period, ensuring that distributed solutions remain physically meaningful.

Results
The framework was tested on the IEEE 33-bus feeder over a 24-hour horizon, partitioned into nine regions.

Curtailment vs. Fairness: The technical benchmark resulted in 2.1097 MWh of renewable curtailment. The fairness-constrained allocation increased curtailment to 5.7216 MWh but capped the maximum cumulative curtailment ratio at 11.20%.
Fairness Indices: The fairness-constrained approach raised Jain indices close to unity (e.g., 0.9985 in nominal conditions), significantly reducing disparity compared to technical-only allocations.
Algorithm Performance: The regional ADMM converged within 208–234 iterations across six different operating conditions (including high PV, high wind, and congestion scenarios).
AC Validation: Independent AC power flow validation confirmed that all solutions remained within the adopted voltage limits (0.90–1.05 p.u.) and thermal limits, with voltage deviations ( $\Delta V^{AC}$ ) below 0.01 p.u.
Storage Impact: BESS effectively alleviated curtailment impacts by shifting midday renewable energy to evening peaks, though some conditions (e.g., peak demand with low renewable availability) still required minor demand response.

Significance and Claims
The paper claims that the proposed approach offers a solid alternative to single-period DOE designs, which are often overly conservative or fail to capture temporal equity. By co-designing multi-period fairness (which inherently increases curtailment to ensure equity) with battery storage (which alleviates the impact of that curtailment), the framework provides a more robust mechanism for modern DOE design. The authors emphasize that this integrated multi-period approach is necessary to properly quantify the cost of fairness and to ensure that distributed solutions are physically valid under AC constraints. The study concludes that separating technical limits, fairness budgets, and storage recourse allows for a transparent trade-off analysis that single-objective OPF embeddings cannot isolate.

Technical Summary: Fair Dynamic Operating Envelopes using Distributed Multi-Period Optimal Power Flow and Jain Index

More like this