⚡ electrical engineering

Allocation of Dynamic Operating Envelopes in Radial Distribution Networks

This paper analyzes how various factors influence dynamic operating envelope (DOE) calculations in radial distribution networks and proposes a novel, transparent, and scalable algorithm called LACE to optimize capacity allocation for distribution system operators.

Original authors: Wilhiam de Carvalho, Florin Capitanescu, Cyril Rasic, Jean-François Toubeau, François Vallée

Published 2026-05-11

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Wilhiam de Carvalho, Florin Capitanescu, Cyril Rasic, Jean-François Toubeau, François Vallée

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 Big Picture: The "Dynamic Envelope"

Imagine a neighborhood power grid as a water pipe system.

The Substation is the main water tank at the top of the hill.
The Houses are the taps.
The Pipes are the wires.

In the old days, the water company (the grid operator) would say, "No matter what, you can only turn your tap on to 10%." This is safe, but it's wasteful. Sometimes the pipes are wide open and empty, but the rule stays strict.

This paper introduces a new idea called a Dynamic Operating Envelope (DOE). Think of this as a smart, adjustable limit on your tap.

If the pipes are full of water (high demand), the limit tightens.
If the pipes are empty (low demand), the limit loosens, letting you use more power (or put more power back into the grid, like solar panels).

The goal is to let people use as much electricity as safely possible without breaking the pipes (overheating) or causing the water pressure to drop too low (voltage issues).

The Problem: How to Share the "Extra" Space?

The paper asks a tricky question: If there is extra space in the grid, who gets to use it?

Should the person closest to the main tank get the most? Or should the person at the very end of the street get a fair share? The paper shows that the answer depends entirely on how you do the math.

The authors tested two main ways of calculating these limits:

The "Simple" Math (Linear Model): This ignores tiny energy losses in the wires. It's like calculating travel time by ignoring traffic jams.
The "Real" Math (Non-Linear Model): This accounts for energy lost as heat in the wires. It's like calculating travel time including traffic jams and friction.

What They Found (The Surprising Results)

1. The "Heat" Constraint (Thermal Limits)

Imagine the pipes can only handle so much water before they burst.

With Simple Math: It doesn't matter who gets the water. The system thinks everyone is equal. You could give the extra space to the house next door or the house at the end of the street, and the total amount of water allowed stays the same.
With Real Math: The location does matter.
- For Import (Using Power): It's better to give the extra space to houses closer to the main tank. Why? Because water traveling a long way loses energy (heat). Closer houses waste less.
- For Export (Putting Power Back): It's actually better to give the extra space to houses farther away! Why? Because when they push power back, the "waste" (heat) in the long wires actually helps absorb some of that power, relieving stress on the main tank.

2. The "Pressure" Constraint (Voltage Limits)

Imagine the water pressure needs to stay within a specific range. If it drops too low, the shower gets weak.

Both Math Methods Agree: In this case, houses closer to the main tank always get the advantage.
Houses far away have a "penalty" because their power has to travel a long distance, which drops the pressure more. So, the system naturally favors the neighbors closest to the source to keep the pressure stable for everyone.

The New Solution: The "LACE" Algorithm

The authors realized that using complex math engines to solve this every 15 minutes is slow and can sometimes crash (like a calculator freezing).

They invented a new tool called LACE (Linear Analytical Calculation of Envelopes).

The Analogy: Instead of running a massive, slow simulation to find the answer, LACE is like a smart checklist. It quickly looks at the grid, identifies the "bottlenecks" (where the pipes are tightest), and hands out the limits in a logical, step-by-step order.
The Benefit: It is incredibly fast, doesn't need expensive software to run, and gives results that are almost identical to the complex methods. It's like switching from a supercomputer to a reliable, instant calculator.

Real-World Testing

The team tested this on:

A tiny 3-house model.
A realistic 8-house neighborhood in Belgium with real solar panels, electric cars, and batteries.
A massive simulation with over 1,000 houses.

The Results:

Speed: LACE was the fastest, often finishing in a fraction of a second, while the complex methods took much longer.
Accuracy: LACE matched the "Simple Math" results perfectly.
Scalability: Even when they simulated a huge network, LACE stayed fast, proving it could work for entire cities, not just small neighborhoods.

The Takeaway

This paper teaches us that how you calculate the limits changes who gets the power.

If you ignore energy loss, you might treat everyone equally, but you might miss efficiency gains.
If you account for energy loss, the "best" spots to put power change depending on whether you are using it or generating it.

The authors provide a new, fast, and transparent tool (LACE) that helps grid operators figure these limits out quickly and fairly, ensuring the grid stays safe while letting people use more renewable energy.

Technical Summary: Allocation of Dynamic Operating Envelopes in Radial Distribution Networks

Problem Statement
The accelerated adoption of Distributed Energy Resources (DERs) challenges the safe and efficient operation of distribution networks. Traditional grid management tools often result in over-conservative restrictions because they fail to capture the fast-changing behavior of modern networks. Dynamic Operating Envelopes (DOEs) have been proposed as a solution to dynamically adjust operating limits for DERs based on variable grid conditions, thereby unlocking non-utilized network capacity. However, the calculation and allocation of these envelopes are highly sensitive to the underlying optimization formulation. Existing research has explored the impact of objective functions, but the specific effects of power flow models (linear vs. non-linear), binding network constraints (thermal vs. voltage), and calculation cases (import vs. export) on allocation outcomes remain insufficiently understood. This lack of understanding can lead to counter-intuitive allocation results, potentially causing over- or under-allocation of capacity to prosumers.

Methodology
The paper conducts an in-depth analysis of DOE formulation using two primary optimization approaches:

NLP-DOE: Utilizes the full AC non-linear power flow equations (branch flow equations), resulting in a non-convex Non-Linear Programming (NLP) problem.
LP-DOE: Utilizes a linearized power flow model that disregards line losses, resulting in a convex Linear Programming (LP) problem.

The authors analyze the allocation of envelopes under two calculation cases:

Import: Maximizing additional consumption ( $p_m \ge 0$ ).
Export: Maximizing additional injection ( $p_m \le 0$ ).

The study examines how these formulations interact with two binding constraints: thermal limits at the feeder head and voltage magnitude limits at the nodes. Theoretical analysis is provided for the linear model, while empirical analysis is used for the non-linear case. Additionally, the authors propose a novel analytical algorithm, LACE (Linear Analytical Calculation of Envelopes), which directly computes DOEs based on the linear model without relying on external solvers.

Key Contributions and Findings

Impact of Power Flow Models on Allocation:
- Thermal Constraints:
  - Linear Model (LP-DOE): When thermal limits are binding, the model shows no locational penalties. The problem is degenerate, meaning any distribution of the total available capacity among nodes yields the same combined envelope value.
  - Non-Linear Model (NLP-DOE): Line losses introduce locational penalties. For import cases, envelopes are preferentially allocated to upstream nodes (shorter electrical distance) to minimize losses. For export cases, envelopes are preferentially allocated to downstream nodes; the reverse power flow creates losses that alleviate stress at the feeder head, allowing for greater total injection.
- Voltage Constraints:
  - In both linear and non-linear models, voltage constraints consistently favor upstream nodes. Nodes further downstream have higher path impedance and voltage sensitivity, resulting in stricter limits. This preference holds true for both import and export cases.
The LACE Algorithm:
The paper proposes LACE as a transparent and scalable alternative to solver-based optimization. The algorithm iteratively identifies the "least penalized" node based on voltage sensitivity and spare capacity, allocating the most conservative value among envelope bounds, thermal capacity, and voltage capacity. It converges rapidly and does not require third-party solvers.
Numerical Simulations:
Simulations were conducted on a conceptual 3-node feeder, a realistic 8-node low-voltage (LV) feeder using anonymized Belgian data (Flobecq dataset), and scalability tests with networks up to 1,000 nodes.
- Results: The LACE algorithm produced results identical to the LP-DOE approach but with significantly shorter computation times.
- Scalability: LACE consistently demonstrated the fastest computation time. While NLP-DOE computation times varied significantly with topology (sometimes faster than LP-DOE in highly branched networks), LACE remained the most efficient.
- Allocation Discrepancies: In thermally constrained scenarios, the NLP-DOE approach yielded unique optimal solutions (favoring specific nodes based on losses), whereas LP-DOE and LACE yielded multiple optimal solutions with the same total capacity but different individual node allocations.

Significance and Claims
The paper claims to provide crucial insights and tools for Distribution System Operators (DSOs), stakeholders, and academics. Its primary significance lies in demonstrating that the choice of optimization formulation (linear vs. non-linear) and the nature of the binding constraint (thermal vs. voltage) drastically alter the spatial allocation of network capacity.

The authors assert that:

Counter-intuitive outcomes exist: Specifically, the non-linear model's handling of line losses can lead to export envelopes being maximized at downstream nodes, a result not captured by linear models.
Fairness policies must be context-aware: Designing fairness policies (e.g., weighting downstream nodes) without considering the binding constraint type (thermal vs. voltage) and the power flow model could exacerbate fairness issues rather than resolve them.
LACE offers a practical tool: The LACE algorithm provides a reliable, solver-free method for DSOs to compute DOEs efficiently, particularly useful for large-scale networks or as a warm start for more complex NLP optimizations.

The paper concludes that understanding these mechanisms is essential to ensure DOE calculations achieve desirable and efficient outcomes, preventing the misallocation of capacity in modern distribution systems.