Global energy demand is gradually increasing every day, and most consumption still relies on fossil fuels [1
]. This results in the rapidly growing overutilization of natural resources and the emission of anthropogenic greenhouse gases (GHGs). The critical issues of anthropogenic climate change, the energy crunch, and the environmental pollution produced by the combustion of fossil fuels are major concerns worldwide [2
]. It is widely accepted that society needs a change in energy production and consumption [3
]. Consequently, it is necessary to replace fossil fuels with renewable energies (i.e., solar, wind, hydropower, etc.) [2
Renewable energy is considered to be the best solution for various energy challenges, for example, the exhaustion of fossil fuels and environmental pollution, since it significantly contributes to environmental protection, commercial growth, and energy safety [4
]. Thus, in the quest for sustainable development, significant efforts have been invested in exploiting renewable energies, with local governments offering incentives for decreasing conventional energy expenses. Among all the available renewable energy sources, solar energy dominates over others in terms of its energy capacity growth rate and low maintenance cost [5
]. Subsequently, solar energy has been accepted as a clean, abundant, readily available, free, environmentally friendly, and economic energy source [6
], without carbon cost [7
]. The installation of photovoltaic panels on the building roofs has various advantages such as efficient utilization of renewable solar energy and distribution of total residential energy consumption that eventually leads to the reduction of CO2
]. In urban areas, building roofs are considered to be more appropriate than building facades as spaces on which to install solar panels [9
]. In addition, solar photovoltaic (PV) technologies offer the most cost-effective investment, partially since such systems need little maintenance to produce electricity free of GHGs [11
]. Renewable solar energy helps in reducing the emission of GHGs, which also supports green building ranking schemes [12
]. Thus, solar energy contributes to buildings’ energy requirements and provides health benefits, environmental stewardship, and more sustainable cities. Earlier related research aimed at identifying and characterizing the reduced manufacturing costs of photovoltaic devices [13
Solar energy potential is an important attribute in identifying the most appropriate building roof surfaces on which to mount solar panels [14
]. The efficient utilization of solar energy and fulfillment of energy demand for both commercial and residential buildings is still a challenge. Therefore, it is important to accurately estimate solar energy potential and examine its distribution over building roofs. This is not simple as it depends on considerable spatial and temporal variations in solar radiation that are significantly affected by several factors [15
], including different roof geometries [16
], inclined surfaces [17
], shadow effects [18
], geographical locations [19
], and topographic features [20
]. Thus far, several models and technologies have been studied to estimate the energy potential of building roofs [21
Light detection and ranging (LiDAR) mapping is one of these techniques [22
], primarily since the elevation data in urban surroundings can be easily attained through airborne LiDAR scans. Secondly, LiDAR-based data acquisition technology permits the quick reorganization of terrain surfaces when estimating building insolation with existing two-dimensional models. Using 2.5D DSM, Carneiro et al. [23
] established a tool to estimate solar energy that is not only appropriate for building roofs but also suitable for the study of facades. Redweik et al. [24
] then suggested a method to estimate solar energy based on the r.sun radiation model, established by Šúri et al. [25
], and integrated this into the open-source Geographic Resources Analysis Support System (GRASS-GIS) [26
]. Despite having larger areas, the result showed that the building facades return lower solar energy than roofs. A method has been presented to estimate solar potential by using graphic processing units (GPU) with measured unified device architecture technology and a LiDAR dataset [19
]. The LiDAR methods significantly improve the energy potential estimation in terms of the anticipated model error [27
], and they have been frequently used in the calculation of solar energy potential in urban settings by incorporating GIS technologies [28
]. Though LiDAR-based techniques are cost-effective and suitable for buildings, there are many restrictions in terms of computation cost and data availability, and they are limited to the comparatively simple structure of building roofs. In order to overcome such limitations, new methodologies are essential for estimating solar energy. A pixel-based method was applied to estimate solar energy on flat roofs [30
], while Li and Liu [31
] applied a pixel-based approach to pitched roofs.
The primary aim of the current study is to develop a procedure for assessing the rooftop solar energy photovoltaic potential over a heterogeneous urban environment. Such an environment may include flat and pitched roof surfaces at different inclinations and in different directions, along with multi-segment roofs on a single building. Due to the complex roof geometry, very high-resolution data, such as ortho-rectified aerial photography (orthophotos), along with LiDAR data, may be an appropriate means for accomplishing the goal. The previously reported studies show that the aspect and slope information plays a very crucial role in solar estimation analysis [27
]. Therefore, detailed aspect and slope maps are essential. Using the products of airborne LiDAR, a new model was developed to create an aspect-slope map. A minor change in slope and its direction over rooftops is noticeable while working on a single map (i.e., aspect-slope map) that could be overlooked while observing individual maps. Although various studies have been reported to estimate solar energy over building roofs but more specifically, to get the building footprints, an object-based method was applied to segment building as an object [32
]. In some studies, building polygons was directly considered to estimate the PV potential [34
]. To automate this tedious task, the present study develops an object-based algorithm to classify all buildings of the study area. Furthermore, the research assumes that in order to accurately obtain the value of the roof’s aspect and slope, an object-based method, using a data fusion technique, should be applied.
In the paper, Section 2
, describes the complete methodology which is implemented to estimate the PV potential on the building rooftops. The section goes through the various steps including study area and data sources information, applied object-based classification algorithm, LiDAR generated DSM used to produce an aspect-slope map, and solar radiation maps. Finally, spatial analysis has performed over building roofs. In Section 3
, results and discussion has been performed. The subsections of Section 3
, contain the classification and accuracy assessment, outcome of LiDAR mapping, validation of generated aspect-slope map, spatio-temporal distribution of solar radiation, and final results of spatial analysis. Finally, the work has been summarized in Section 4
To fulfill the energy demands of a growing population, the solar panels are a promising source. Therefore, it is necessary to install urban solar panels precisely in the appropriate positions to utilize maximum solar radiation effectively. For efficient solar power exploitation, the sun rays must be perpendicular to the solar panel. Therefore, accurate information about the slope and the aspect of the solar panel is required. This research study helps to find the aspect and slope of building rooftops that provide the maximum solar potential throughout the year. Previously, little research focused on building roofs using object-based procedures along with LiDAR data. Therefore, by capitalizing on established methods, this research has developed an object-based method, including orthophoto, which helps to classify and map building rooftops of various sizes and shapes, while airborne LiDAR systems provide support to find the heights, slopes, and orientations of the corresponding building roof structures. The developed method enables the estimation of solar energy yields on flat and pitched roof surfaces, as well as multi-segment rooftops, through defining solar irradiances in units of pixels over a specific period.
The Kiryat Malakhi area in Israel was selected as the study area because of its infrastructure, comprising one to three floor buildings, with different roof types, flat and pitched, in diverse slopes and aspects along with multi-segment shapes. The area was analyzed based on the aspect-slope map. ESRI’s Solar Analyst toolbox was used to produce solar maps for four specific days of the year and to measure the corresponding solar insolation output for individual grid cells. Finally, the instantaneous solar radiation was calculated at specific times, and the generated statistics are visualized graphically.
The results demonstrate that the applied method of calculation can achieve exact aspects and slopes at which to install solar panels on building roofs that will deliver high energy throughout the year, aiding in the effective utilization of renewable energy sources. The resultant solar insolation maps of specific days demonstrate the high energy throughout the year. Moreover, the generated aspect-slope map gives the optimal slope and direction for installing solar panels on building roofs that provide high solar energy.
The study used direct and diffuse solar radiation, not reflected radiation in the measurement. However, the reflected measurement could be used to obtain more precise insolation maps of building roofs (if the roof surface contains different types of materials). The lack of a cloud cover analysis is another key limitation of the current study. Still, the proposed method can be applied to more accurate solar panel sizing for installation and the precise calculation of solar radiation for tenants and commercial energy investors, resulting in the effective consumption of renewable energy sources. The developed algorithm can be applied to areas where rooftop information is required. In summary, the study can be expanded to estimate solar insolation on the facades of buildings, and future research can be undertaken for particular applications of solar devices.