aqreed/solarpy

Solar radiation model for flight dynamics. Based on Duffie & Beckman "Solar energy thermal processes" (1974)

Description	Python Solar Radiation model
Author	aqreed [email protected]
Version	0.1.3
Python Version	3.6
Requires	Numpy, Matplotlib

This packages aims to provide a reliable solar radiation model, mainly based on the work of Duffie, J.A., and Beckman, W. A., 1974, "Solar energy thermal processes".

The main purpose is to generate a solar beam irradiance (W/m2) prediction on:

• any plane, thanks to the calculation of the solar vector in NED (North East Down) coordinates, suitable for its use in flight dynamics simulations...
• any place of the earth, taking into account the solar time wrt the standard time, geometric altitude, the latitude influence on solar azimuth and solar altitude as well as sunset/sunrise time and hour angle, etc.
• any day of the year, taking into account the variations of the extraterrestrial radiation, the equation of time, the declination, etc., throughout the year

Example 1

Solar irradiance on the southern hemisphere on October 17, at sea-level 13.01UTC (plane pointing upwards)?

import numpy as np
from solarpy import irradiance_on_plane
from datetime import datetime

vnorm = np.array([0, 0, -1])  # plane pointing zenith
h = 0  # sea-level
date = datetime(2019, 10, 17, 13, 1)  # year, month, day, hour, minute
lat = -23.5  # southern hemisphere

irradiance_on_plane(vnorm, h, date, lat)

A dedicated Jupyter Notebook on beam irradiance can be found here.

Example 2

Power output (in W) of a solar panel with the following characteristics:

• surface of 2.1 sqm
• efficiency of 0.2
• pointing upwards
• in NYC
• on December 25, at 16.15

from numpy import array
from solarpy import solar_panel
from datetime import datetime

panel = solar_panel(2.1, 0.2, id_name='NYC_xmas')  # surface, efficiency and name
panel.set_orientation(array([0, 0, -1]))  # upwards
panel.set_position(40.73, -73.93, 0)  # NYC latitude, longitude, altitude
panel.set_datetime(datetime(2019, 12, 25, 16, 15))  # Christmas Day!
panel.power()

Example 3

Solar declination on August 5?

from solarpy import declination
from datetime import datetime

date = datetime(2019, 8, 5)  # August 5

declination(date)

Please find more notebooks on the 'examples' folder that you can open locally, or just try to launch online interactive Jupyter notebooks.

---

NOTE: solarpy is under development and might change in the near future.

---

Dependencies

This package depends on Python, NumPy and Matplotlib and is usually tested on Linux with the following versions:

Python 3.6, NumPy 1.16, Matplotlib 3.0

Installation

solarpy has been written in Python3, and its version v0.1 is available in PyPi. It can be installed using:

$ pip install solarpy

To install in development mode:

$ git clone https://github.com/aqreed/solarpy.git
$ cd solarpy
$ pip install -e .

Testing

solarpy recommends py.test for running the test suite. Running from the top directory:

$ pytest

To test coverage (also from the top directory):

$ pytest --cov

Bug reporting

Please feel free to open an issue on GitHub!

License

MIT (see COPYING)

solarpy's Issues

add diffuse radiation calculation

until now, the irradiance_on_plane function only calculates direct radiation. Diffuse radiation might account for as much as 15% of the total radiation on a clear-sky or 100% in a cloudy day.

https://en.wikipedia.org/wiki/Diffuse_sky_radiation

It must be valid for altitudes ranging from 0 to 25km

add a DEM (digital elevation model)

this would allow to take into account local shadows

• https://pypi.org/project/elevation/#description
• http://geologyandpython.com/dem-processing.html

use datetime objects

functions shall accept datetime objetcs instead day_of_the_year/month/day/hour/minute, in order to make the code cleaner

allow the definition of geographical points of interest

Instead of latitude + solar/standard time...Find a way of enabling the use of longitude to define a city for example

Underestimation by a factor of 2

I tried to adapt the example panel_class.ipynb to a typical solar panel and its mounting position. My impression is that I have to multiply all estimated results for the panel power by two to fit to the measured results.

404 on Raspberry OS

pip install solarpy results in Environment error Client error 404

add a weather prediction module

right now solarpy uses a clear-sky model, which is not accurate.

modified both direct beam and diffuse radiation due to weather