A General PV Digital Twin Concept: Streamlining the PV Plant Lifecycle Data Exchange

Introduction

The PV industry is rapidly advancing in digital transformation. Digital twins are increasingly used throughout the lifecycle of a PV asset, from design and monitoring to decommissioning. However, challenges remain in replicating a physical PV asset digitally and achieving uniformity across the lifecycle. Current practices involve manual assignments, which are labor-intensive and potentially inaccurate, and are repeated at each lifecycle stage with different requirements and information sources. This results in multiple independent digital twins tailored to specific processes, complicating transitions between lifecycle stages due to incompatibility. This report is based on the TRUST-PV Horizon 2020 funded research project, with partners working to develop commercially viable digital solutions.

A General Digital Twin for PV Plants

Digital twins support PV performance simulation by providing information to performance models, enabling simulation of yield, degradation, and reliability. These models facilitate optimal decision-making throughout a PV plant's lifetime, from design and procurement to operation and maintenance (O&M).

2.1 Definition of Digital Twin

For solar PV systems in TRUST-PV, a digital twin is defined as:

A parametrized (2D/3D) model of a PV system that contains all the physical information needed to simulate the behavior and performance of the real PV plant it represents.

2.2 Digital Twin of a PV Plant

The physical information needed to simulate a real PV plant can be broken down into:

1. 1.3D geospatial context: Includes sources of shading, such as:
   • ​Far-shading from terrain features (e.g., mountains)
   • ​Near-shading from local objects (e.g., trees, buildings)
2. 2.3D geometric properties of the active PV faces (solar modules)
3. 3.Electrical hierarchy: How components are connected to form a functional system
4. 4.Electrical properties of individual components (e.g., PV modules, inverters)

2.3 General Digital Twin Requirements

A general digital twin should enable collaboration across multiple platforms and lifecycle stages. Key features include:

• ​3D model with all modules, shading objects, and their optical properties
• ​Change tracking with a versioning system
• ​Component metadata (e.g., serial numbers, geolocation) for unique identification and BIM compatibility
• ​Complete electrical design and component characteristics, provided with high accuracy at delivery

Creating PV Plant Digital Twins – The Role of PV Engineering Software

PV design software translates ideas, data, and best practices into PV plant design files. Advanced tools like PVcase Ground Mount generate structured information models that can be reused for digital twins. The report explores both new plant design and remodeling of existing plants using incomplete design files and documentation.

3.1 New Plant in Design Phase

When designing a new PV plant, it's important to use software with an information model compatible with the general digital twin requirements. Design files can then be used to create digital twins, with efficient versioning for design iterations. Tools like PVcase Ground Mount support 3D topography-based layout and electrical design, helping achieve capacity targets, manage shading, and implement framing and stringing practices, while considering civil engineering aspects.

3.2 Existing Plant 3D Layout

For existing plants, matching real entities with digital ones is challenging. Two approaches are:

• ​Drone survey: Scans modules, terrain, and shading objects in 3D. Post-processed data is used in design software to complete the digital twin, independent of as-built documentation.
• ​Existing documentation: Uses plant metadata, 2D locations, and geometry of PV frames. More accurate models are possible with terrain topography data from site surveys. Shading objects can be included via drone data or remodeled based on known location and extent.

3.3 Existing Plant Electrical Design

Accurate documentation is crucial for digitizing the electrical topology of existing PV systems. While automated parameter identification is being developed, human interpretation of electrical diagrams is currently necessary due to lack of standardization. Design tools like PVcase Ground Mount automate the incorporation of electrical design and 3D cable paths, essential for accurate performance simulations. Component specifications are sourced from datasheets or manufacturer files.

Demonstration in a Real Scenario

The Don Rodrigo 2 PV plant (Sevilla, Spain, BayWa r.e.) serves as a case study for digital twin creation for performance simulations. The 50 MWp plant, commissioned in 2020, has 131,404 PV modules on HSATs with North-South axis, connected to string inverters.

4.1 As-Built Documentation and Processing

Access to as-built CAD design files simplifies digital twin creation. Key steps include:

• ​Identifying important layers and objects: terrain, frame polygons, cable paths, inverters, substations, shading objects
• ​Assigning topography as terrain surface
• ​Converting frame polygons to PVcase Ground Mount objects by setting geometry attributes
• ​Adapting frames to terrain topography automatically
• ​Replicating electrical design by reading single-line diagrams (SLDs) or CAD files, defining stringing patterns, and assigning components
• ​Re-creating cabling by defining trench lines and using automated cable generation algorithms
• ​Adding 3D cables for accurate length values
• ​Re-creating shading objects (e.g., transformer stations) based on design files

4.2 Digital Twin from Drone Survey Data

Drone-survey-based digital twin creation is similar to documentation-based reconstruction, using CAD output with ground topography, frame objects, and shading objects. Differences include:

• ​Drone data only includes visible objects; cable trench locations must be estimated
• ​Provides accurate 3D representation of objects and shading
• ​Reduces dependence on documentation
• ​Electrical connectivity still requires manual interpretation of SLDs or CAD layers

4.3 Yield Simulation Using the Digital Twin

The digital twin can be used as input for physics-based performance simulation software (e.g., IMEC's framework, PVcase Yield). Simulations comparing drone-based and as-built models for a single inverter showed similar results, indicating comparable accuracy.

Executive Summary

This report summarizes the results of TRUST-PV Task 2.1, establishing a PV plant digital twin definition compatible with all lifecycle processes. The digital twin acts as a single source of static asset information. New requirements improve digital representation of 3D geometry, geospatial context, and functional connectivity, increasing similarity between physical and digital assets. Methods enable accurate 3D modeling and far shading reconstruction using high-resolution elevation models. The main challenge remains digitalizing electrical design and component identification, which are not visible to optical sensing and rely on documentation.

The new digital twin concept supports advanced, physics-based yield simulations, mapping results to the 3D model for location- and context-dependent analysis. The same digital twin structure can be used throughout the lifecycle: design, optimization, monitoring, O&M, decision support, and BIM. The demonstration case shows practical feasibility and accuracy benefits of sensing the real asset. Future work should focus on automating electrical connectivity discovery, validating simulation frameworks, and integrating with advanced digital services. Benchmarking digital twin accuracy at monitoring and inspection stages will help evaluate the concepts' value. WP7 of TRUST-PV will attempt practical implementation to assess technical and economic feasibility.

Industrial Implementation of the Developed Concept

PVcase Ground Mount and Yield enable fast and efficient digital twin construction. Engineers can design digital twins and export them for yield simulation. Both tools can convert drone survey data into digital twins of existing plants.

The digital twin concept enables accurate simulations and serves as a communication tool between digital processes such as system design, yield simulation, O&M, and construction monitoring.

The ultimate benefit is providing a single source of truth about physical asset information from early design to decommissioning, facilitating information federation and data-informed decision-making.