As Distributed Energy Resources (DER) continue to proliferate across UK transmission and distribution networks, understanding curtailment risk has become critical for network operators, developers and investors alike. Curtailment estimates underpin connection decisions, valuation and long‑term investment confidence, yet the modelling methods that derive these estimates, the Curtailment Assessment, are not always well understood.

Smarter Grid Solutions has been modelling DER curtailment impacts for over sixteen years, using time step analysis to study the time-varying impact of grid constraints and the resulting curtailment actions. Our consultants are expert in modelling and assessing network behaviours including impacts from markets, regulatory change, future demand and generation uptake. We study how these factors can impact DER developments looking to connect to transmission or distribution utility networks. Some clients are asking SGS to look at evolving impacts over 10-to-20-year horizons considering planned network changes and the connection timelines for DER accepted-but-not-connected from the Embedded Capacity Register (ECR).

This article explores the different modelling approaches that can be applied to connection curtailment assessments, their relative strengths and limitations, and when simplified methods may be appropriate for representative early‑stage analysis, particularly as networks, markets and regulatory frameworks continue to evolve. We feel it is important that DER developers and investors have an understanding of the different approaches and of those, what are most appropriate based upon the nature of curtailment, complexity of the network, or the stage of development for the site.

For time-series curtailment studies, there are typically two general analysis approaches accepted for modelling curtailment:

 1. Arithmetic Data Driven Network Models  

Data-driven linearised models can offer fast and efficient curtailment assessments with acceptable accuracy, though this accuracy depends on the complexity of the network case and availability of data. SGS often uses this method to consider constraints at a single GSP or even when operating in parallel with other GSP networks which requires information on the sensitivity factors relating DER to each GSP.

This method forms the basis of some of the DNO curtailment assessment tools available on their data portals. In these cases, ‘offline’ load flow analysis is carried out for multiple network topologies, producing power flow and sensitivity factor datasets that form the inputs to these web-hosted, data-driven analysis tools. The method benefits from a straightforward analysis process, providing a fast-to-implement and fast-to-simulate approach to curtailment assessment.

Limitations of this method include:

• The models are dependent on the availability of suitable data, in particular the historical underlying power flows through the constrained circuits of interest. Ideally the DNO will publish the historical power flows required and most UK DNOs publish datasets of historical power flows through their network Grid Supply Point (GSP) transformers. For circuits downstream of the GSP however, power-flow datasets are not always available from the DNO data portal. While there are methods to estimate these power flows, these will be less accurate than measured historical data and for non-radial networks the level of accuracy in the power flow estimates may not be suitable to ensure representative curtailment estimates.
  
  
• Future DER and demand profiles are combined with the historical power flows to simulate future power flows where the impact of DER on the circuit of interest can be modelled using a known linear relationship, e.g. a sensitivity factor. On radial networks, sensitivity factors are straight forward to estimate, however, on more complex interconnected or meshed networks, without a suitable load flow model you are dependent on the DNO making this information available.
  

 2. Load Flow Network Models 

In some cases, curtailment assessments cannot rely on historical power flow data or pre‑calculated sensitivity factors. This may be because suitable data is not available from the DNO, or because the network itself is sufficiently complex that estimating power flows would introduce unacceptable uncertainty. In these situations, a load flow in-the-loop modelling approach becomes essential.

With this method, a full load flow model of the relevant network is used to simulate network power flows, performing this simulation at each assessment time step. For every time step, power flows and voltages are calculated and checked against defined constraint thresholds at the network asset. Where a constraint threshold is exceeded, a representation of the network control system such as an Active Network Management (ANM) scheme determines the appropriate control actions. The load flow simulation is then re‑run under the curtailment scenario to verify that network constraints are removed and to quantify the resulting curtailment applied to each affected DER site. Repeating this process across all time steps allows curtailment to be aggregated over the study period and reported as energy (MWh) or financial impact.

Load flow in-the-loop assessments use full, non‑linear AC power flow equations, enabling accurate simulation of both thermal and voltage constraints. This significantly improves study fidelity compared with purely data‑driven approaches, particularly in meshed networks, at electrically distant constraint locations, or where voltage‑driven ANM behaviour is material. In these scenarios, the approach provides a more complete and defensible assessment of curtailment risk.

Where up‑to‑date load flow models are readily available, the timescales and efficiencies of this approach can be comparable to data‑driven methods. If DNO load flow models require updating, some additional effort is needed, but delivery timelines remain practical for most development and investment use cases. The most resource‑intensive scenario arises when no suitable model exists and a network representation must be built from first principles however, this can still be justified where accuracy is critical to commercial or strategic decisions.

What Does this Mean for Developers and Investors?

Selecting the right modelling approach for curtailment assessments is important to ensure the accuracy achieved suits the project development stage and decision-making process. In our experience, SGS will always assess the best approach based on client requirements, knowledge of the DNO networks and our access to DNO data and load flow models.

In addition to the modelling approach, other data and assumptions have significance such as DER production profiles and associated influencing factors including weather trends, markets behaviours and regulatory rules. SGS has worked with many clients to bring curtailment modelling closer towards an operational reality and away from the conservative assumptions typically implemented in network design activities.

SGS is deeply engaged across the electricity sector, providing analysis and advisory support to stakeholders from developers and investors in renewable energy systems to DSOs and Transmission Owners/Operators. For many years we also have supplied software systems for energy asset control and delivered specialist advisory services such as constraint and curtailment modelling. We closely follow industry changes relating to the connection and management of diverse generation and storage technologies. It is at the confluence of these various activities that we have informed the ongoing development of our curtailment assessment processes.