Optimising virtual site assessment process for mounted PV array design: an immersive BIM-based approach for existing buildings

Purpose The assessment and performance evaluation of photovoltaic (PV) systems depend heavily on the site characteristics of existing buildings. However, when physical site inspections are impractical or impossible due to restricted access, this creates a significant barrier to the assessment of PV layout design. This research uses an instrumental case-study methodology to optimise PV array design in existing buildings by integrating Building Information Modelling (BIM) with immersive technologies. Design/methodology/approach A two-stage optimisation approach was developed to address challenges associated with physical site assessments and fixed building locations. By utilising optimisation algorithms, the grasshopper optimisation algorithm, and Unreal Engine 5 (UE5) for immersive visualisation, this method facilitates detailed shadow analysis and optimal placement of fixed-mounted PV panels. The methodology was validated through a case study involving the modelling of an office building in Melbourne, Australia, using Revit. A rigorous comparative analysis of the developed workflow and the Solar Engineer tool was conducted on the same building to assess the efficiency and accuracy of the workflow. Findings Results demonstrated that integrating BIM with immersive technologies substantially enhanced the efficiency and accuracy of PV installation assessments. The analysis identified a roof utilisation factor (Roof UF) of 39.62% for PV panels, enabling the installation of 90 solar panels with an annual energy yield of 152,091.64 kWh. Originality/value The optimisation workflow introduced in the paper highlights its potential to reduce the reliance on physical site visits while improving design precision and maximising energy output. These findings provide valuable insights for solar engineers and building professionals, contributing to the advancement of efficient and sustainable urban energy solutions.