RenewML_Energy

Survey of machine learning methods to predict renewable energy grid load and output.

Data Module

• Data()
Base object for data loading and analyzing.
• Data.get_data(datafile: str, powerfile: str=None, dropna: bool=True, rescale_power: bool=True)
Method to retrieve and format data from CSV.
• datafile (str): location of 4y historical weather data file.
• powerfile (str): location of power output file from solar stations.
• rescale_power (bool, default: true): Whether to rescale power from W to kW.
• Data.transform(lag: list or str, resample: str=None, scaler: str=None, copy: bool=True)
Make input data stationary
• lag: Input list of choice from {15minutes|minute|day|week|month|season|year} Input choice of {day|week|month|season|year} for differencing or list of multiple for multi-order differencing.
• scaler: Use scikit-learns scaler to standardize or normalize data. choose from [ minmax | standard ]
• resample: Numpy frequency string. Frequency to resample data to. Default frequency is inferred from powerfile.

VAR model