Open meljsingh opened 3 years ago
Workflow of forecasting model
in Python
- Download latest data in R:
- Clean latest data:
- take hourly average
- make sure temperature and energy data is aligned (clock change)
- construct lights algorithm
- identify date and time of forecast
- create training dataset (1 year)
- Fit model:
- BSTS dynamic model
- seasonal ARIMA static model
- Predict temperatures
- output mean forecast, upper and lower bounds for each model
- output probability too hot/ too cold
- Save outputs in format readable by Python
This model is in the models folder here
To dos (in order of priority
Aim is to add temperature forecasting model onto platform, visualise outputs, and provide relevant warnings.
Explanation of the model
Output table for 1 sensor location:
In addition, there are the following four outputs:
Further information:
Further analysis
The error from the previous day's forecast can provide further information