Having developed a workable initial dashboard for testing purposes using Plotly I have now explored the weather data from Keele using DEOPS. The data is as expected in different formats depending on whether you connect to the API or do a manual download. For the speed of developing a test pipeline, I have started with a manual download.
I have had a big choice to make which will affect the accuracy of the final model. Keele have only been collecting Solar production data consistently for about 6 months. Prior to that, there are some quite big gaps in the data, which means I have either 12 months of data with big gaps or 6 months with no gaps. As my final product will be tested with September / October 2021 forecasts I feel it is important to have data, even incomplete data, going back to September 2020. I have yet to decide how to deal with the missing data.