PVGIS TMY POA Irradiance Model

Description:

This repo contains the code required to obtain TMY data from PVGIS for any location in the world. This is then used to calculate from the total plane of array (POA) irradiance in kWh/m2 striking a modelled surface. The only requirements for this program to calculate TMY and POA data is to input a Latitude and Longitude to indentify location, however more relevant parameters can be adjusted (see below).

This is useful when conducting any modelling which requires incident radiation values, and calculates values within a 1 hour resolution. This script can be run by importing the packages into an appropriate IDE, or by running the PVGIS_irradiance.py file from wtihin this directory. The parameters that can be adjusted are as follow:

See included Jupyter Notebook (PVGIS.irradiance.ipynb) for step by step breakdown on how to utilise code.

Modelling Parameters:

• latitude: Latitude of the site (Default: 54.60452)
• longitude: Longitude of the site (Default: -5.92860)
• surface_pitch: Pitch of the modelled plane in degrees (Default: 35)
• surface_azimuth: Azimuth of the modelled plane in degrees (Default: 0)
• albedo: Albedo of the modelled plane (Default: 0.2)
• refraction_index: Refraction index of the modelled plane (Default: 0.1)

Example Chart showing grouped weekly radiation values produced from model

Currently once intialised the Sites class will collect annual TMY data from the EU JRC PVGIS API, it will then perform calculations to determine POA radiation, which will then be grouped and stored within the class object. The full time series modelled radiation data can then be accessed directly in timeseries format with 8760 (1 hour) entries, or alternatively the grouped and aggreagated data can be accessed for summaried information.

See uploaded Jupyter Notebook for step by step method accessing of data. Also see folder of example_charts where I have included some examples of charts made from directly accessing the grouped radiation values. This was done in the included Jupyter notebook file if you would like to see the methods used.

Example CLI commands to run PVGIS_irradiance.py:

Full Command

python PVGIS_irradiance.py --latitude 54.60452 --longitude -5.92860 --surface_pitch 35 --surface_azimuth 0 --albedo 0.2 --refraction_index 0.1

Simple Command

python PVGIS_irradiance.py --latitude 54.60452 --longitude -5.92860

Current implementation of model:

The model is created by initializing a Site class object when the program is run, which then conducts the appropriate modelling. The TMY and Irradiance data simulated can then be accessed using .xxx notation. These models are returned as DataFrames, with the library being built with a focus on using it offline for personal data analysis.

Example Jupyter Notebook Implementation

from meteo.Site import Site

latitude = 54.60452
longitude = -5.92860
surface_pitch = 35
surface_azimuth = 0
albedo = 0.2
refraction_index = 0.1

site = Site(latitude, longitude, surface_pitch, surface_azimuth, 
            albedo, refraction_index)

Accessing model results:

# Example on how to access summary of model results
site_tmy_data = site.tmy_data
site_irrad_data = site.irrad_model

site_irrad_data_hourly = site.rad_data.hourly
site_irrad_data_daily = site.rad_data.daily
site_irrad_data_weekly = site.rad_data.weekly
site_irrad_data_monthly = site.rad_data.monthly
site_irrad_data_quarterly = site.rad_data.quarterly

Saving model results:

# Example on how to access summary of model results
import pandas as pd

site_irrad_data_daily = site.rad_data.daily

site_irrad_data_daily.to_csv("Testy_Test_Name.csv")

This is an example implementation using basic inputs, however there are many more options for customising the model by adding additional inputs. All of the data is stored within pandas dataframes, and as such allow for the usual methods of indexing and accessing data within these structures.

Current Model Classes & Methods:

Site

site = Site()
site.latitude
site.longitude
site.surface_pitch
site.surface_azimuth
site.albedo
site.refraction_index

site.tmy_data
site.irrad_model

site.rad_data.hourly
site.rad_data.daily
site.rad_data.weekly
site.rad_data.monthly
site.rad_data.quarterly

Requirements:

pip install httpx
pip install numpy
pip install pandas