Skip to main content

Tools for performing common tasks on solar PV data signals

Project description



Explore our documentation · Report Issue

Latest Release latest release
License license
Build Status documentation build status Actions build status
Publications DOI
PyPI Downloads PyPI downloads
Conda Downloads conda-forge downloads
Test-Coverage test-coverage

Solar Data Tools is an open-source Python library for analyzing PV power (and irradiance) time-series data. It was developed to enable analysis of unlabeled PV data, i.e. with no model, no meteorological data, and no performance index required, by taking a statistical signal processing approach in the algorithms used in the package’s main data processing pipeline. Solar Data Tools empowers PV system fleet owners or operators to analyze system performance a hundred times faster even when they only have access to the most basic data stream—power output of the system.

Solar Data Tools provides methods for data I/O, cleaning, filtering, plotting, and analysis. These methods are largely automated and require little to no input from the user regardless of system type—from utility tracking systems to multi-pitch rooftop systems. Head over to our Getting Started pages in our documentation for a demo! For an in-depth tutorial on Solar Data Tools, we recommend taking a look at the recent webinar we did with the DOE's Solar Energy Technologies Office (SETO) with our colleagues at NREL, linked below:


You can also check the notebooks folder in this repo for more examples.

This work is supported by the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy (EERE) under the Solar Energy Technologies Office Award Number 38529.

Install & Setup

Recommended: Install with pip

In a fresh Python virtual environment, simply run:

$ pip install solar-data-tools

or if you would like to use MOSEK, install the optional dependency as well:

$ pip install "solar-data-tools[mosek]"

Install with conda

[!WARNING] When installing solar-data-tools using conda, you will need to add three channels, conda-forge, slacgismo, and stanfordcvxgrp, to your conda config (or alternatively specify them using the -c flag as shown in the examples below). Failure to do so will result in the installation of an outdated solar-data-tools version. Note that we will be moving solar-data-tools to conda-forge soon, which will simplify the installation process. Check back soon for an update! For more on conda channels, see the conda docs.

Creating the environment and directly installing the package and its dependencies from the appropriate conda channels:

$ conda create -n pvi-user solar-data-tools -c conda-forge -c slacgismo -c stanfordcvxgrp 

Starting the environment:

$ conda activate pvi-user

Stopping the environment:

$ conda deactivate

Or alternatively install the package in an already existing environment:

$ conda install solar-data-tools -c conda-forge -c slacgismo -c stanfordcvxgrp 

Solvers

CLARABEL

By default, the CLARABEL solver is used to solve the signal decomposition problems. CLARABEL (as well as other solvers) is compatible with OSD, the modeling language used to solve signal decomposition problems in Solar Data Tools. Both are open source and are dependencies of Solar Data Tools.

MOSEK

MOSEK is a commercial software package. Since it is more stable and offers faster solve times, we provide continuing support for it (with signal decomposition problem formulations using CVXPY). However, you will still need to obtain a license. If installing with pip, you can install the optional MOSEK dependency by running pip install "solar-data-tools[mosek]". If installing from conda, you will have to manually install MOSEK if you desire to use it as conda does not support optional dependencies like pip.

More information about MOSEK and how to obtain a license is available here:

Usage

Users will primarily interact with this software through the DataHandler class. By default, Solar Data Tools uses CLARABEL as the solver all signal decomposition problems. If you would like to specify another solver (such as MOSEK), just pass the keyword argument solver to DataHandler.pipeline with the solver of choice.

from solardatatools import DataHandler
from solardatatools.dataio import get_pvdaq_data

pv_system_data = get_pvdaq_data(sysid=35, api_key='DEMO_KEY', year=[2011, 2012, 2013])

dh = DataHandler(pv_system_data)
dh.run_pipeline(power_col='dc_power')

If everything is working correctly, you should see a run summary like the following

total time: 25.99 seconds
--------------------------------
Breakdown
--------------------------------
Preprocessing              6.76s
Cleaning                   0.41s
Filtering/Summarizing      18.83s
    Data quality           0.21s
    Clear day detect       0.44s
    Clipping detect        15.51s
    Capacity change detect 2.67s

You can also find more in-depth tutorials and guides in our documentation.

Contributing

We welcome contributions of any form! Please see our Contribution Guidelines for more information.

Citing Solar Data Tools

If you use Solar Data Tools in your research, please cite:

Recommended citations

Bennet E. Meyers, Elpiniki Apostolaki-Iosifidou and Laura Schelhas, "Solar Data Tools: Automatic Solar Data Processing Pipeline," 2020 47th IEEE Photovoltaic Specialists Conference (PVSC), Calgary, AB, Canada, 2020, pp. 0655-0656, doi: 10.1109/PVSC45281.2020.9300847.

Bennet E. Meyers, Sara A. Miskovich, Duncan Ragsdale, Mitchell Victoriano, Aramis Dufour, Nimish Telang, Nimish Yadav, Elpiniki Apostolaki-Iosifidou, Claire Berschauer, Chengcheng Ding, Jonathan Goncalves, Victor-Haoyang Lian, Tristan Lin, Alejandro Londono-Hurtado, Junlin Luo, Xiao Ming, David Jose Florez Rodriguez, Derin Serbetcioglu, Shixian Sheng, Jose St Louis, Tadatoshi Takahashi, and Haoxi Zhang. (2024). slacgismo/solar-data-tools. Zenodo. doi: 10.5281/zenodo.5056959

Citing technical details (e.g., SDT algorithms)

Bennet E. Meyers, PVInsight (Final Technical Report). United States. https://doi.org/10.2172/1897181

Citing a specific version

You can also cite the DOI corresponding to the specific version of Solar Data Tools that you used. Solar Data Tools DOIs are listed at here.

Versioning

We use Semantic Versioning for versioning. For the versions available, see the tags on this repository.

Authors

See also the list of contributors who participated in this project.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

solar_data_tools-1.6.2.tar.gz (77.0 MB view hashes)

Uploaded Source

Built Distribution

solar_data_tools-1.6.2-py3-none-any.whl (77.0 MB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page