Mapping Future Energy

Editor's Notes

• #2: I am Adrian Moisey, I am the GIS Officer for National Grid Electric Transmission, I have been working for NG now for 20 years and have been working with GIS systems for the last 10 years and FME for the last 7 years.
• #4: In 2008 we moved from using the old OS LandLine product to OS MasterMap, as part of that transition we were given an FME workbench by Intergraph that managed part of the process for loading the data and change only updates into our Oracle database. Shortly after starting to use FME for this we discovered it was capable of doing much more than just helping us load OSMM data and could be a very powerful tool to the team, allowing us to automate tasks we undertook on a daily basis, as well allowing us to preform queries for the business in a quick and efficient manner. In 2013 we put together a business case to purchase FME Server, this has enabled us to share the power of FME to the wider business without the necessity of training people how to use FME Desktop, using FME server has improved efficiency within the GIS team and has also allowed the business to self-server data without having to come to our team for it.
• #6: Besides the original reason for purchasing FME it is now one of our key tools within the team. Some of the things we have used it for are merging spatial and non-spatial data sets together (something I will talk about in a bit more detail later on). Spatially representing priority work on our Electricity Transmission network, we are currently trying to increase the throughput of work on the network and being able to see the work on a map has allowed the business to make better decisions on what work can be grouped together to improve efficiency. We use FME to generate Tiff images using OS OnDemand, when we moved to Vector Map Local OS gave us a workbench to rasterise the VML data, however it wasn’t perfect and performance was not great so I looked at using FME to do the job for us. We also use FME for handling file updates received from OS, this includes things like copying the files and moving them to the desired location as well as updating the data in our Oracle database. We also use FME for creating queries and output for the business, if it is something the business wants to perform on a regular basis then we also make this available on FME Server so they can self-serve the data without coming to us every time the need it updating.
• #8: SO had millions of records for Solar PV (on peoples roofs), Electric Vehicle charging points as well as wind and solar farm locations. These would be assigned a general location using a very adhoc method based on a number of assumptions, for instance the residential model they used was based on wealth and house size. Prior to this work they had to use an A0 map of the UK to get a general idea of the location. The data was not granular enough for their customers needs, the customers needed the data at a higher level of detail down to smaller operation zones, this just was not available to business.
• #9: SO had been working with Oxford University to generate the smaller operational zones required, this work had come to an end and needed to be taken over by ourselves. The data supplied for the locations of Solar PV/wind farms etc was all spreadsheet based and contained no coordinates for us to easily georeference against, the only data supplied we could use was a post code or partial post code, we therefore used this to create a spatial reference for each row of data. I used FME to match the data provided to a postcode and then generated a spatial data set from this, through running through several iterations and scenarios in the workbench I was able to spatially reference most of the data. The resulting data sets were then output to a spreadsheet for easy reference outside the GIS and also made available to users within the business via our GIS system.
• #10: This is one of the workbenches I created to spatially reference the data sets SO provided me, you can see a number of scenarios being stepped through to attempt to match as much of the data as possible. I could possibly have created a custom transformer in here to do the matching, maybe something I can do to improve the workbench in the future.
• #11: Here is an example of the outputs I generated for SO. The blue polygons you can see are the smaller operation zones they needed to map the data against, this is the level of detail their customers required but they could not provide previously. The other data shown here is the location of plugin hybrid car charging points, this was provided as a text file from the DVLA.
• #12: Here we can see where the solar PV installations are around the country, the results of creating this data set were actually a surprise for SO, they were expecting the locations to be more dispersed around the country, where as the data visualised on a map shows it is more concentrated around the south of the country.
• #13: This slide shows us the locations of Hydro generation, again the data set when visualized against a map surprised SO, they had based their models on them being located more in the north of the country as opposed to being more widely dispersed as this shows.
• #14: SO now have a much more accurate view of where these features are located, as I mentioned previously, instances of solar installations are heavily concentrated towards the south of the country than their models had indicated and the hydro generations is much more dispersed than they thought. The information they hold on these is now much more detailed, the locations of these features allows the business to make much better informed decisions and has helped improve their processes.