Resource Forecasting Algorithm

Abstract

=============================================================================================== Resource Forecasting Abdul Motin HOwlader and Hamed Mohsenian-Rad University of California, Riverside July 2019 ===============================================================================================

---------------------------------4h-ahead of prediction----------------------------------------

Python File: Forecast_CE-CERT_4h_Pred.py

To run this Python File: Python Version 3.6.3 and Keras API are required.

Deep neural network method such as Long short-term memory (LSTM) was applied to forecast the PV power.

The Python File reads all the input parameters from the "CE-CERT_1200_4h.csv" for LSTM network.

Output Results: 4h-ahead of PV power forecasting and it will be saved in the "file_path.csv" file. An actual and prediction graph will be displayed. Percentage normalize root mean square error (%nRMSE) will be shown as an output result.

In output "file_path.csv", the first column refers as number of data point, second column for actual data, and third column for prediction data. ---------------------------------24h-ahead of prediction----------------------------------------

Python File: Forecast_CE-CERT_24h_Pred.py

To run this Python File: Python Version 3.6.3 and Keras API are required.

Deep neural network method such as Long short-term memory (LSTM) was applied to forecast the PV power.

The Python File reads all the input parameters from the "CE-CERT_1200_24h.csv" for LSTM network.

Output Results: 24h-ahead of PV power forecasting and it will be saved in the "file_path.csv" file. An actual and prediction graph will be displayed. Percentage normalize root mean square error (%nRMSE) will be shown as an output result.

In output "file_path.csv", the first column refers as number of data point, second column for actual data, and third column for prediction data.