Photovoltaic: Where is it Best?

Where solar photovoltaics make the most sense

Notebook ►Pitch

Challenge #10

(english below)


Die Verbreitung der Solar-Photovoltaik in der Schweiz ist nur in grossem Umfang möglich, wenn sie wirtschaftlich sinnvoll ist und gleichzeitig die Treibhausgasemissionen reduziert.


Diese beiden Faktoren können durch einen Vergleich der Lebenszykluskosten (LCC) und der Lebenszyklustreibhausgasemissionen (LCE) mit ihrem lokalen Stromtarif und dem LCE des Strom-mixes der Versorgungsunternehmen bewertet werden. Die LCC und die LCE variieren für jedes einzelne PV-System, abhängig von ihrer Grösse, den erneuerbaren Ressourcen und anderen technischen Faktoren.

Angesichts der bisherigen Erfahrungen des PSI über das PV-Erzeugungs-potenzial und -kosten kann diese Analyse durchgeführt werden, um mehr Erkenntnisse darüber zu gewinnen, wo der Einsatz von Solarphotovoltaik in der Schweiz am sinnvollsten ist.

Where solar photovoltaics make the most sense

The big picture:

The rollout of solar photovoltaics to a great extent in Switzerland is only possible when they make economic sense and at the same time reduce greenhouse gas emissions.


These two factors can be assessed by comparing the Life Cycle Cost (LCC) and Life Cycle greenhouse gas Emissions (LCE), with their local electricity tariff and the LCE of electricity mix supplied by utility providers. The LCC and LCE varies for each individual PV system, depending on their size, renewable resources, and other technical factors.

Given the previous experience on PV costs and generation potential, this analysis can be undertaken to give more insights on where it makes the most sense to deploy solar photovoltaics in Switzerland.

Call for expertise

  1. GIS expert
  2. Application/UI developer
  3. Energy consultant/scientist
  4. Expert from government/utilities working on PV system rollout/implementation

Questions to answer in two perspectives:

  1. Country-wide (mainly analysis): how much potential do we have for Switzerland considering LCC and LCE in a realistic sense

  2. Individual resident (requires UI):

    • Does it make sense to install PV on top of my roof
    • How can I find residents around me with similar conditions/interest to reduce system investment cost (economy of scale)


  1. levelized cost of electricity for each roof in Switzerland from this analysis by Technology Assessment Group, Laboratory for Energy Systems Anlaysis, PSI
  2. local electricity mix by electricity generation technology for municipalities and entities in 2018, obtained from Stromkennzeichnung
  3. 2 sources for PV potential in Switzerland
  4. life cycle greenhouse gas emissions (LCE) per kWh for different electricity generation technologies based on ecoinvent version 3.6
  5. electricity tariffs in Switzerland from ElCom
  6. simulating the performance of photovoltaic energy systems (pvlib-python)
    • is a community supported tool that provides a set of functions and classes for simulating the performance of photovoltaic energy systems. It was originally ported from the PVLIB MATLAB toolbox developed at Sandia National Laboratories.
    • see more on project gibhub page
  7. remuneration from electricity providers

Thinking process

In this project we did additional study and compiled research notes, sample data and suggestions, expanding on the original challenge

  1. Limitations of the calculator

    • 1.1 Number of factors: it considers factors like solar irradiance, angle of roof area, shading. The new study also considers additional factors like temperature (which affects the module efficiency), obstacle structure on roof (based on machine learning) = more realistic estimate --> use data from new study
    • 1.2 Calculation not based on current technology/efficiency factor: Someone planning to build a new PV installation would (reasonably) use the latest technology, while most calculations by tools are based on averages (which consist of a mix of old and new technology). Estimates for kWh potential might be 1/3 to 1/2 too pessimistic. --> use better data based on new technologies?
    • 1.3 Grid network restrictions (maximum power which can be fed into the network): depending on location (more rural = usually worse), maximum capacity which can be fed into the network is much lower, dependent on three factors: connecting cable ampacities, voltage range limitation, and transformer power rating. Data on these three factors is proprietary information of the electricity companies. Might be in a range between 0 and 90% restriction on potential --> data is not openly available. Is there any way we can find a proxy? (especially for CH-wide potential estimate) --> for individuals data might be retrievable from their energy provider for inclusion in calculators (e.g BKW online portal offers the information for individuals for their own place)
  2. Calculate potential (economical)

    • 2.1 Consider energy usage in network: network limitations might be less extreme because numbers given are based on no consumption --> unclear how this can be estimated
    • 2.2 effects of everyone having PV installed: excess energy supply at peak times in an area = limitation of the network
  3. Calculate potential (environmental)

    • 3.1 data on energy mix by city is available

Link to Drawing of data structure

Network Limitations

The past generation potential estimates don't seem to take into account the limitations of the network infrastructure. There are three limiting factors related to adding PV to the existing network: connecting cable ampacities, voltage range limitation, and transformer power rating:

  • the voltage increase due to feeding in from a new PV cannot exceed +3% threshold allowed by law
  • the current increase due to feeding in from a new PV cannot cause the current to exceed a cable rating (on the entire chain from house to transformer)
  • the power fed in by a new PV cannot exceed the transformer rating; the ones connected to the medium voltage grid or others connecting any intermediate subtransmission (1000V intermediate voltage system) network

The available public data does not include any of this (cable ratings, network topology, transformer ratings) since it is all proprietary to the power company (Elektrizitätsversorgungsunternehmen or EVU).

Using proprietary data and comparing the Sonnendach available roof area capacity (sum of all roof sections identified, i.e. good, marginal and bad, using the maximum area) with the calculated maximum feed in values for the electrical connection of the utility, shows a large overestimate of the energy capacity for rural and industrial sites.

Location Type Sonnendach Network Max % Over
Uerzlikon house 75500 59000 28
Uerzlikon house 61700 71000 0
Uerzlikon apt 2½ 56800 48000 18
Uerzlikon farm 128300 29750 331
Uerzlikon farm 505400 33850 1393
Biel industrial 235400 183000 29
Biel industrial 244300 140000 75

The network maximum is calculated under the worst case assumption that all PV (proposed and installed) are generating at the maximum (peak) value and no consumption is occuring. By and large, the majority of these limitations are based on an over-voltage situation (>3% above nominal). This engineering limit would not normally apply when the PV owner self-consumes or the energy is shared with neighboring consumers.

These network maximums are also calculated when only one additional PV is installed (at the indicated site). So this brings up the additional issue that, should everyone in an transformer service area have a PV installation, these maximum values would be replaced by other limitations, for example overloading of shared cables, overloading of supplying transformers or general over-voltage conditions from multiple injection points. A more detailed analysis is needed on a per transformer service area basis with perhaps PV installations at only the most opportune locations (large area, south facing roofs).

Thus, the “Life Cycle Cost”, LCC, would need to include upgrades to the network infrastructure with the addition of large amounts of PV, or the maximum potential for PV installation would need to be adjusted downwards under the assumption of no network upgrades.

Regarding the proprietary data needed for the engineering analysis:

  • the data is owned separately by each of the over two hundred EVU in Switzerland
  • these EVU are very unlikely to be able to provide this network data due to privacy restrictions and competitive considerations
  • only a small fraction of the German speaking EVU are doing this maximum feed in analysis on a broad scale
  • for these few, the best approach may be to provide a list of candidate sites with kW suggested and ask if their network can support this size of installation at those sites

Launched at Energy Hackdays 2020 by

nikki_böhler Zuzfil xiaojin.zhang julius_chrobak

Maintainer nikki_böhler

Updated 31.08.2020 08:15

  • xiaojin.zhang / update / 28.08.2020 08:33
  • xiaojin.zhang / update / 28.08.2020 08:20
  • xiaojin.zhang / update / 28.08.2020 08:19
  • xiaojin.zhang / update / 28.08.2020 08:18
  • xiaojin.zhang / update / 28.08.2020 08:17
All attendees, sponsors, partners, volunteers and staff at our hackathon are required to agree with the Hack Code of Conduct. Organisers will enforce this code throughout the event. We expect cooperation from all participants to ensure a safe environment for everybody. For more details on how the event is run, see the Guidelines on our wiki.

Creative Commons LicenceThe contents of this website, unless otherwise stated, are licensed under a Creative Commons Attribution 4.0 International License.