Big Solar Co-op is developing tools to help volunteer solar activists to identify and assess potential community-owned solar roofs. A subset of those tools will involve GIS. Where possible we want to re-use existing technology but we are also prepared to develop our own.
Our broad approach is as follows:
- Site identification
- Outline Site screening. If positive then:
- Detailed Site assessment 1
- Approach building owner. If positive then:
- Detailed Site assessment 2
- Sign up building owner. If signed up then:
- Solar design and installation
There are several layers of site assessment here:
Site identification. At the moment we are bringing together sites which have been suggested by community solar activists. In the future we want to be able to be more systematic in the way that we identify potential sites.
Outline site screening. May be carried out by a range of different people:
- Volunteer activists
- Site owners themselves
- Paid staff in some cases
The assessment aims to quickly establish how viable for solar a site is. This involves the collection of data relating to solar PV itself and data about the site owners, electricity users etc. The focus is on collecting data and establishing which sites have a decent chance of being viable – this does not have to be 100% accurate.
Big Solar Co-op is developing its own online tool to enable a range of people to quickly assess sites. Version 1 of this tool will only have limited solar analysis, in order to make it easy to use – at around the same detail level as broad-brush tools like PVGIS (indeed we will probably integrate the simple PVGIS API in this process).
Detailed site assessment. Will be carried out by either trained volunteer activists or paid staff. The aim of this stage is to create a preliminary solar design which enables us to model the site in more detail:
- PV output (kWH over a year amount and distribution)
- Layout scenarios
- Identification of site-specific issues and opportunities
This is the kind of assessment which is commonly made by sales teams in commercial solar operations. At this level there is a good range of software available. Big Solar Co-op is currently using OpenSolar and is beginning to train volunteers in its use. OpenSolar was chosen as it is:
- Free at point of use
- User-friendly and does not require a large amount of solar expertise
- Detailed enough to enable a proposal to be made to building owners
Solar design for a signed-up site will be done by paid experts, initially as part of the installation contract and as time goes on more probably by a dedicated team within Big Solar Co-op. Designers might use a tool like PVSOL which offers 3D design to the level of module strings, cable runs etc.
Site identification is the main area where there are gaps in provision.
Community groups often wish to assess the solar potential of a town, village or region. Frequently consultants are brought in to conduct a GIS-based assessment which typically results in a list of sites ordered by solar output potential.
There are 2 problems with these reports. Firstly, it is often not clear what to do next! Presenting them to the Local Authority is seldom likely to result in action. We want to use this kind of report to build local action. A local group working with Big Solar Co-op can use this as a source of leads. It could also make the data public as a way to promote its work and the wider solar agenda locally. It would help if the report output formats were standardised across the UK.
Secondly, the reports tend to be costly, only done where there is funding and not easily replicated, updated or expanded. We have already had some success in replicating consultants reports using open-source software and freely-available datasets. We think there is a niche for a new product here – a freely-available web service enabling communtiy solar activists to identify suitable roofs in a given area.
The level of analysis could improve over time. As a bare minimum we can identify:
- Roof area
- Insolation (from lat/long)
- Half-hourly output predictions
With the use of LIDAR data we can compute roof slope, azimuth and approximate shading in most places allowing considerable improvements in assessment.
There are multiple non-PV datasets which could be linked to provide potentially useful site-related data:
- Land Registry and Companies House for identification of possible businesses on site
- Cadastral datasets provided by public bodies
- Grid constraint maps provided by DNSO/DSOs
The minimum viable product is probably a cookbook enabling a moderately technically capable user to conduct an outline analysis using QGIS – explaining where to get datasets, how to set up the solar analysis tools etc.
From there we want to move to create a much more user-friendly interface. This product would probably have the following features:
- An automated script enabling creating of a list of sites in standard format from a given bounding box of co-ordinates
- Utilisation of open-source software such as QGIS, Openlayers etc
- Utilisation of existing algorithms such as those provided by SEBE, SAGA or GRASS (r.sun)
- A dedicated server running the necessary software for analysis and reporting, with automated download of updated data on a regular basis
- An open code base hosted on gitlab
- A pluggable architecture allowing community developers to add features and datasets
- An interactive interface allowing selection of a region and changing settings
- A non-interactive interface (simple API/REST)
The API would enable us to enhance the Outline Site Screening procedure for a single site. In this configuration a web service request could be made to enable a better broad-brush assessment than is currently possible, and to return any useful metadata held on the system.
Big Solar Co-op is confident that it can obtain the core funding needed for servers, and any crucial software/data licences. We are seeking to put together a team of GIS users and coders to create a basic codebase and to build a small community of code contributors, maintainers, trainers and advocates. These roles will initially not be paid though there will be long-term paid server and codebase maintenance roles.