Evergreen: A Renewable Energy Simulator

Introduction

# "`-''-/").___..--''"`-._
#  (`6_ 6  )   `-.  (     ).`-.__.`)   WE ARE ...
#  (_Y_.)'  ._   )  `._ `. ``-..-'    PENN STATE!
#    _ ..`--'_..-_/  /--'_.' ,'
#  (il),-''  (li),'  ((!.-'
# 
# Authors: 
#     Weiming Hu <weiming@psu.edu>
#     Guido Cervone <cervone@psu.edu>
#
# Geoinformatics and Earth Observation Laboratory (http://geolab.psu.edu)
# Department of Geography and Institute for CyberScience
# The Pennsylvania State University

This project is designed for assessing prediction uncertainty for renewable energy production. It builds on top of Analog Ensemble and pvlib. It currently supports running with data from Analog Ensemble and from SURFRAD.

It is implemented in Python 3.

Usage

For your reference, here provides a collection of scripts to prepare the environment on HPC platforms.

Once you have the environment set up, you can use the following code to see the available options:

python runner_pv_anen.py -h

python runner_pv_surfrad.py -h

Profiling

yappi

# Run the program through the profiler
python runner_pv.py --profile --profiler yappi

# On Mac OS. kcachegrind on Linux
qcachegrind yappi_2020-04-12-18-08-48_rank-0.log

pyinstrument

# Run the program through the profiler
python runner_pv.py --profile --profiler pyinstrument

line_profiler

# Run the program through the profiler
kernprof -l runner_pv.py --profile --profiler line_profiler

# Generate text output
python -m line_profiler evergreen.py.lprof

Simple Clock

If you are ready for production but still want to have a general idea of how much time was spent, the simple clock would be a good solution. It just uses time.time() to get current time so its overhead is pretty low.

python runner_pv.py --profile

Feedback

We appreciate collaborations and feedback from users. Please contact maintainer Weiming Hu or create tickets if you have any problems.

Thank you!