The project aims to enhance Distribution Network Operator (DNO) network modelling capabilities to make them scalable and future proof. This involves developing additional features and integrating the required datasets. If successful, the project will enable network modelling use cases including functionality to flexibly isolate sub-networks, state estimation, and integration of forecasts.
Once complete, this open-source tool will enable members of the public to run power flow simulations using published models. This project will offer accessibility to network modelling capabilities for customers supporting innovation across the sector. Python Power will support the transition to Net Zero through better use of data and open innovation.
Benefits
The Python Power project is about enabling smarter, faster, and more flexible electricity network modelling using open-source tools. By adapting the widely used PandaPower platform (open source), the project allows UK Power Networks to simulate and analyse power flows more efficiently, helping to forecast constraints, support planning decisions, and improve operational responsiveness. The project supports innovation, transparency, and long-term cost savings by removing licencing barriers and encouraging in-house tool development.
The main quantifiable benefit which should be realised on completion of the project and implementation into BAU operation will be an increase in flexibility utilisation through more accurate forecasting with PandaPower, and the ability to estimate the impact of flexibility dispatch on constraints more robustly
Learnings
Outcomes
The project successfully delivered a Python-based modelling capability that enables UK Power Networks’ network models to be imported, analysed, and used for power flow studies using open-source tools.
Key outcomes include improved modelling flexibility (e.g. sub-network analysis), enhanced understanding of data and modelling requirements, and the demonstration that open-source tools can be interoperable with DNO network models. The learning generated provides a foundation for future adoption of open-source network modelling across the industry.
The project has delivered clear benefits for customers by enabling UK Power Networks’ network models to be used with open‑source tools, removing the need for proprietary software. These network models have been released in the publicly available LTDS Common Information Model (CIM) format, alongside similar releases by other DNOs, increasing accessibility and standardisation across the industry.
This allows a wider range of users (including academics, developers, local authorities, and innovators) to analyse network constraints, test scenarios, and identify viable connection opportunities. As a result, customers benefit from improved planning confidence, more transparent and usable network data, ultimately supporting better network utilisation and lower costs.
Lessons Learnt
The learning generated from this project can significantly benefit network licensees by demonstrating that open-source tools can be demonstrated to be interoperable with existing network models. This project demonstrated how to develop and implement a flexible, programmatic approach to power flow simulations using Python. The open-source nature of this software will allow for continuous improvement and customisation, enabling network operators to adapt the tool to their specific needs.
A key lesson learned relates to the complexity of working with real-world network models and ensuring they perform as expected within a new modelling environment. While initial assumptions were that models could be readily transferred and analysed, the project identified numerous unexpected challenges, including variations in how assets are represented, differences in modelling approaches, and the need to accommodate more complex equipment such as three-winding transformers and station controllers. Significant effort was required to diagnose and resolve issues affecting power flow convergence and model stability. This highlights the importance of early data validation, iterative testing, and close collaboration between domain experts and developers in future projects to ensure robust and scalable solutions.
