Renewable energy potential: evaluation for Switzerland

Buildings represent one of the largest shares of the energy demand in Switzerland: they account for more than 40% of the overall energy demand and more than 30% of the electricity demand. To reduce energy consumption as well as greenhouse gas emissions, buildings need to become much more energy-efficient and rely primarily on renewable energy resources. Accordingly, estimating a HyREP for the built environment is an important issue in Switzerland, particularly for municipalities, building owners and public utilities. Such estimates provide very useful information regarding renewable energy generation as well as their potential for satisfying the energy demand in the built environment.

Hybrid renewable energy systems combine solar, wind power and shallow geothermal energy. In the building sector, such systems substantially reduce the size of standalone power plants, the required energy storage capacity, and overall operating costs. In view of the energy transition, it is important to assess the potential of combined renewable energy resources in the built environment to support stakeholders’ decisions and policies for the corresponding sector in Switzerland.

This project aimed to estimate the hybrid renewable energy potential (HyREP) in the built environment at national scale, with application to Switzerland. The latter describes the spatial and temporal patterns of several renewable energy (RE) resources for electricity and heat generation. They may be combined with other non-renewable energy sources to assess complementarities between them and their potential for satisfying the energy demand in the built environment. The renewable energy resources addressed in this work were wind power, solar photovoltaic (PV) electricity and shallow geothermal heat. These are three renewable energy resources with ambitious expansion goals in the Swiss Energy Perspectives for 2050.

Using Big Data approaches, the project allowed to assess how much electricity and heat may be generated throughout Switzerland from solar, wind and shallow geothermal energy. The results may impact energy policies in this country, and the data-driven methods can also be used for other countries. The renewable energy database allows us to visualise the spatio-temporal variation of the renewable energy resources for individual buildings as well as for municipalities anywhere in Switzerland.

The focus of the project lies in the spatial and temporal estimation of the technical potential of RE sources at large scale. The technical potential is defined as the maximum energy (electricity or heat) which can be extracted using a specific renewable energy technology. Its estimation accounts for physical, geographical, and technical constraints, which were addressed separately for all three forms of energy (wind, solar and shallow geothermal). The results include regional and national databases of RE potential, which can be used to study hybrid renewable energy systems from neighbourhood to country scale based on data that are homogeneous across Switzerland.

A generic data driven methodology, based on Big Data mining, spatial statistics, and machine learning, so as to favour the high-resolution modelling of hybrid (solar, wind, and geothermal) RE systems were set-up. Software packages containing the developed models for further applications and extensions are available. The methodology for the estimate of hybrid renewable energy potential has been applied to the building stock in Switzerland. This allowed to compare hybrid renewable energy potentials for different urban and building typologies (for example rural vs urban or residential vs industrial), as well as for different Swiss regions, and to derive useful conclusion for further development of renewables. Uncertainties for the estimated and forecasted values (confidence levels and prediction errors) have been determined and illustrated by means of interactive maps and visualisation tools for decision-making.

Hybrid renewable energy potential for the built environment using big data: forecasting and uncertainty estimation