2019 FOSS4G Bucharest Talks speaker: Christian Braun


Talks

A Scalable Approach for Spatio-Temporal Assessment of Photovoltaic Electricity Potentials for Building Façades of Entire Cities

The assessment of renewable energy potentials in urban environments gained a lot of interest in the recent decades due to CO2 reduction goals by cities, national policies as well as directives by the EU. In combination with advances in data creation and processing as well as the definition of standards like CityGML, new ways of modeling urban potentials have been developed. This lead to numerous approaches estimating roof-top solar photovoltaic (PV) production. However, in recent years due to research in building materials, the façades became more attractive and feasible for PV electricity production.
This paper describes results on the development of an completely FOSS-based approach to assess the electricity production potential by building façade PV. To estimate solar irradiation we followed the hemispherical viewshed approach described by Fu, 1999. Combining it with an approach to dissect walls into regular 3D point grids (1 meter spacing) we calculate the sun visibility (each hour) and the sky viewshed throughout the year. This results in direct and diffuse irradiation for every wall point. To generate the electricity potential, the irradiation values are summed up for the wall points and are fed into an economic model. This is driven by technical parameters of the installation, such as module efficiency, installation and maintenance costs, figures about payback tariffs and envisaged module lifetime.
The overall result is a city-wide PV suitability and electricity production potential map of every building façade.
The processing is based on a city model in the CityGML format using the 3DCityDB database and the spatial processing functionalities of PostGIS. A set of Python scripts has been developed as a central control instance. The scripts control the processing of direct and diffuse irradiation as well as clear sky irradiation relying on the external “pvlib” Python library. Furthermore, we use the scripts to manage parallel processing of queries against the database to achieve scalability and improved performance. The parallelisation is done by processing single building walls. We run a case study with approximately 7000 single wall elements to process. We identified so far one of the major bottlenecks of the approach. This are the calculations of sun visibility for every wall point per timestamp (intersection with surrounding buildings) which takes per wall several minutes to process depending on the number of points per wall.
Since we implemented a parallel processing of the walls running on a 80-core dedicated server machine, the completion for an entire city of 3 million wall points uses a decent amount of time for the given size of data set. Here we describe a scalable and highly parallelised approach which can be easily implemented through standard tools and libraries. This open up now for distributed approaches using multiple database servers for even better scalability.

A Tensor Based Framework For Large Scale Spatio-Temporal Raster Data Processing

In this paper, we address the course of dimensionality and scalability issues while managing vast volumes of multidimensional raster data in the renewable energy modeling process in an appropriate spatial and temporal context. Tensor representation provides a convenient way to capture inter-dependencies along multiple dimensions. In this direction, we propose a sophisticated approach of handling large-scale multi-layered spatio-temporal data, adopted for raster-based geographic information systems (GIS). Moreover, it can serve as an extension of map algebra to multiple dimensions for spatio-temporal data processing. We use the multidimensional tensor framework to model such problems and apply computational graphs for efficient execution of calculation processes. In this approach, spatio-temporal data can be represented as non-overlapping, regular tiles of 2-D raster data, stacked according to the time of data captured. As a case study, we quantify the spatio-temporal dynamics of solar irradiation calculations and 2.5-D shadow calculations for cities at very high space-time resolution using the proposed framework. For that, we chose Tensorflow, an open source software library developed by Google using data flow graphs and the tensor data structure. We provide a comprehensive performance evaluation of the proposed model against r.sun based on GRASS GIS. Benchmarking shows that the tensor-based approach outperforms r.sun by up to 60%, concerning overall execution time for high-resolution datasets and fine-grained time intervals for daily sums of solar irradiation [Wh.m-2.day-1]. Precisely, the main characteristics of the proposed framework include defining, optimizing and efficiently calculating mathematical expressions involving multi-dimensional arrays (tensors); Transparent use of GPU computing such that the same code can be run either on CPUs or GPUs; Implicit parallelism and distributed execution with high scalability offered by data-flow based implementation. Moreover, the Python implementation of the proposed model makes it GRASS GIS ‘Add-on’ compatible.