The Digital Twin: a game-changer in PV design

A digital twin is an emerging technology in the PV industry, offering vast capabilities for measuring PV plant performance and supporting optimal engineering decisions. The following information is based on the TRUST-PV Horizon 2020-funded research project.

What is a digital twin?

A digital twin is a digital representation of a physical object or system. In the context of PV (photovoltaic) plants, it is a parametrized (2D/3D) model containing all the information needed to simulate the behavior and performance of the real PV plant it represents.

One of the primary uses of digital twins is PV performance simulation. The digital twin's information can be used for PV-performance modeling, helping to achieve optimal decision-making by simulating the plant's behavior.

The key data areas required for a PV digital twin include:

• ​3D geospatial context: A 3D representation of the PV plant's location, considering terrain, mountains, and other features that may cause far-shading or near-shading. Accurate representation of these sources is crucial for assessing their impact on performance.
• ​Geometric properties of PV modules: Detailed information about the 3D geometric properties of the solar modules, including positioning, orientation, and alignment.
• ​Electrical hierarchy: The electrical design and hierarchy, detailing how components are connected, including modules, inverters, and other electrical parts.
• ​Electrical properties of components: Data on efficiency, power ratings, electrical characteristics, and other relevant properties of PV modules and inverters.

To ensure effective use throughout the PV plant's lifecycle, a digital twin should have:

• ​A 3D model including PV modules, shading objects, and optical properties
• ​Terrain topography information
• ​Detailed skyline description for accurate far shading modeling
• ​Complete electrical design and component characteristics
• ​Component metadata (e.g., serial numbers, geolocation) for unique identification and compatibility with systems like BIM
• ​Support for changes and versioning
• ​A standardized and open format for compatibility and integration across platforms

Creating a PV plant digital twin requires specialized PV engineering software, such as PVcase Ground Mount, which can generate design files and structured information models. During the design phase, using compatible design software is crucial for meeting digital twin requirements. Advanced PV design tools assist with 3D layout, topography-based design, electrical design, and more, resulting in detailed and accurate 3D models.

For existing PV plants, two approaches can be used to create the digital twin:

• ​Comprehensive drone survey: Conducting a drone survey to scan modules, terrain, and shading objects in 3D. The data is post-processed and combined with PV design software to complete the digital twin, reducing error propagation from documentation.
• ​Using existing documentation: If drone surveys are not feasible, existing documentation (metadata, 2D locations, geometry, terrain data) can be used. PVcase Ground Mount software can automate much of this process.

Well-maintained documentation is crucial for the electrical design of existing PV plants. Human interpretation of electrical diagrams can digitize the electrical topology, and PV design tools can automate the incorporation of electrical design into the model, including 3D cable paths. Component datasheets from plant documentation are the primary source for installed electrical component specifications.

By integrating data from various sources with PV engineering software, a comprehensive digital twin can be created for simulation, analysis, and collaboration across all lifecycle stages.

PVcase case study

Two main approaches to creating the digital twin were used in a case study:

As-built documentation-based digital twin creation

• ​Identify and extract essential layers and objects from as-built CAD design files (terrain topography, frame polygons, cable paths, inverters, substations, shading objects).
• ​Assign terrain topography as the ground surface to adapt frames to local ground topography.
• ​Convert frame polygons into PVcase Ground Mount objects, adjusted to follow terrain.
• ​For electrical design, read single-line diagrams or CAD files to define stringing patterns, assign strings to inverters, and inverters to substations using PVcase Ground Mount's toolbox.
• ​Re-create cabling by locating original cable trench lines and defining them as trench line objects. Automatic algorithms generate DC and AC cable paths.
• ​Manually re-create shading objects based on design files.
• ​Finalized digital twin can be visually inspected and exported in various formats.

Drone survey data-based digital twin creation

• ​Drone survey data is delivered in CAD format, containing ground topography, frame objects, and shading objects as mesh objects.
• ​The 3D representation enables more accurate reconstruction.
• ​Cable trench locations may need to be estimated, as drone data only includes visible objects.
• ​Shading objects can be accurately captured and matched with pre-defined shapes.

Yield simulation using the digital twin

Once the digital twin is created, it can be used for physics-based PV system performance simulation using software such as IMEC's energy yield simulation framework or PVcase Yield. Simulation results of a single inverter based on drone-based and as-built models showed good agreement, indicating the accuracy of the drone-survey-based model for simulation purposes.

Benefits and future developments

The digital twin concept enables advanced, physics-based yield simulations that consider complex 3D illumination and detailed electrical connectivity. It can be used throughout the PV plant lifecycle, from design to operation and maintenance, and serves as a communication tool between digital processes.

Future developments should focus on automating the discovery of electrical connectivity and component characteristics, validating the accuracy of automated digital twin-based yield simulation frameworks, and integrating them into advanced digital services. Benchmarking the accuracy and efficiency of different PV digital twins at monitoring and inspection stages could help evaluate their value.

The case study demonstrates the practical feasibility and applicability of these concepts, paving the way for industrial implementation in the PV industry.