TMY weather files: building energy demand-based generation parameters
The following database contains TMY weather files generated based on the methodology outlined in "Defining generation parameters with an adaptable data-driven approach to construct typical meteorological year weather files" currently submitted to the journal of Energy and Buildings. The TMY weather files are generated using the Sandia Laboratory method for which the generation parameters have been defined using gradient boosted tree regression model. The monthly weighting factors are obtained using hourly building energy demand for a medium office building. The file naming standard is as follows: Climate Zone_Station Number_Application Type_System_Algorithm-Weighting Factor Frequency Montreal weather file example: "CZ6_7025251_MediumOffice_IAL_XGB-MH_WF.epw" CZ6: Climate Zone 6 7025251: Station Number (CWEEDS) MediumOffice: Regression model trained using building energy demand obtained for a medium office building IAL: The system is based on "Ideal Air Load" XGB: The selected algorithm is XGBoost MH: Monthly (M) weighting factors are obtained and the model is trained using hourly data (H)
Steps to reproduce
The steps to reproduce are outlined in the "Defining generation parameters with an adaptable data-driven approach to construct typical meteorological year weather files" which has been submitted to the journal of Energy and Buildings.
Natural Sciences and Engineering Research Council of Canada