Wesim data

Description of Wesim data

The Wesim data is static and read in once at the beginning of the visualisation.

Notes / open questions:

• It takes the form of an Excel spreadsheet with multiple sheets.
• From a parsing perspective, is the starting cell of each table on each consistent, or should some slightly more intelligent detection of where each table starts be employed?
• We probably need a dictionary of the region keys: {"SCO": "Scotland", "NEW": "North England and Wales", "MID": "Midlands", "LON": "London", "SEW": "South England and Wales"}
• All but "Capacity" are hourly data and there can be multiple tables per sheet, separated by a blank column.
• All hourly data tables (in the sample data at least) cover the same timeframe (1 - 168).
• As far as I can tell, every value in the spreadsheet is in MW.
• Values on order of 10^2 - 10^4 in sample data.
• Some of the technologies in "Capacity" are the names of tables in the sheets for hourly data (e.g. onshore wind, offshore wind, solar pv..) but they don't all have a table.

Capacity

2D grid of technology vs Region. Suggest pandas dataframe indexed by technology:

              | scotland | North Eng&Wal | ...
onshore wind  | 16, 525  | ...
offshore wind | 2,770    |
...

Hourly data

For each of the tables listed, a pandas dataframe indexed by hour and with columns ["SCO", "NEW", "MID", "LON", "SEW", "Total"]

RES output

• Onshore wind
• Offshore wind
• Solar PV
• Hydro
• Other RES

Storage output

• Battery Storage
• Pumped hydro storage

Demand

• Demand

Interconnector flows

This sheet is the odd one out. The table is still indexed by hour but column headings represent interconnectors. There is a separate table which acts as a key. The key could probably just be stored as a dictionary.

Hour | SCO-IE | NEW-NOR | NEW-IE | SEW-CE | Total
1    |   491  | 1,238   | 491    | 1,170  | 1,426
2    | ...
...

Key:

Cod    |  capacity  | Notes     |
SCO-IE | 500        | Moyle interconnector (Scotland - Northern Ireland)
...

Interaction with Opal data

We are assuming the WESIM data is completely static

• Keep a copy on the server
• Yue's code doesn't automatically process it, so requires manual changes in the model if there is a change to the data