Solar Photovoltaic Resource Assessment From The Macro Scale To The Meso Scale Using Airborne Laser Scanning Data And Open Source Tools
An understanding of the urban local potential (roof space and solar exposure) is critical for utility planning, accommodating grid capacity, deploying financing schemes and formulating future adaptive policies to take advantage of PV technology's continued price declines. This thesis provides a sequence of methodologies to acquire and pre-process data and access solar PV potential from i) a regional scale, ii) a municipal scale and iii) a municipal district unit. This research explores and confirms the robustness and versatility of the r.sun module within the open source Geographic Resources Analysis Support System (GRASS) for the purpose of solar irradiation modeling at different levels of surface complexity. At the regional scale, the fourteen counties in southeastern Ontario that served as a case study was found to possess over 935,000 acres appropriate for solar farm development, which could provide 90 GW of PV. At the municipal scale, another methodology was developed to provide urban solar PV resource assessments, which is widely applicable throughout the world. The results of the case study indicate that utilities needing to plan for large scale solar electric generation in urban areas can make a rule of thumb estimate. In the absence of advanced computational expertise and high quality remotely sensed data, 30% of the projected area of roofprints can be used as a first pass estimation of the available area for PV installation. Finally, for the municipal district unit, a separate workflow that attains a 95% accurate segmentation from raw and randomly chosen Light Detection And Ranging (LiDAR) data was demonstrated. The solar PV resource assessment on this scale returned 22,000m2 of roof area out of a two dimensional extent of approximately 40,000m2 in building coverage as suitable for the purpose of solar PV installations. With higher integration of urban texture, it was found that the global horizontal irradiation is lower than the regional scale by 30% due to near shadowing effects in downtown Kingston. The three methodologies presented here at three scales provide the foundation to enable automated analysis of solar PV potential and planning for any given region with marginal LiDAR data.