optimPV: Optimization & Modeling tools for PV research

Authors

Vincent M. Le Corre
Larry Lüer

Institution

CAPE - Centre for Advanced Photovoltaics and Thin-film Energy Devices, University of Southern Denmark, Denmark

Description

This repository contains the code to run optimPV. optimPV combines sevral optimization procedures and modeling utilities that can be used for two objectives:

1. to optimize the parameters of a simulation to fit experimental data.
2. to optimize the processing conditions in a self-driving experimental set-up.

Repository Folder Structure

.
├── Main                             # Main directory
    ├── optimPV                      # Main directory for the optimPV codes
        ├── axBOtorch                # Directory with the Bayesian optimization (BO) codes using BoTorch and Ax
        ├── BayesInfEmcee            # Directory with the Bayesian inference codes using emcee
        ├── DDfits                   # Directory with the different agents to run the drift-diffusion simulator  [SIMsalabim](https://github.com/kostergroup/SIMsalabim) for JV, Hysteresis, Impedance, CV and IMPS simulations and fitting
        ├── Diodefits                # Directory with the agent to simulate and fit the non-ideal diode equation model
        ├── general                  # Directory with general utility functions used by the different agents
        ├── posterior                # Directory with some utility functions to plot the posterior distributions using the BO surrogate model
        ├── RateEqfits               # Directory with the agent to simulate and fit the rate equations for different experiment types (trPL, TRMC, TAS, etc.)
        ├── scipyOpti                # Directory with the optimization codes using scipy.optimize
        ├── TransferMatrix           # Directory with the agent to run the transfer matrix simulations
    ├── Notebooks                    # Contains clean versions of the Notebooks
    ├── Data                         # Contains some example data for the notebooks
    ├── docs                         # Contains the documentation
    ├── test                         # Contains the codes for testing optimPV
└── README.md

Installation

With pip

To install optimPV with pip you have two options:

1. Install optimPV using the PyPI repository

   pip install optimpv
   

2. Install optimPV using the GitHub repository https://github.com/openPV-lab/optimPV

   pip install git+https://github.com/openPV-lab/optimPV
   

With conda

To install optimPV with conda:

conda create -n optimpv 
conda activate optimpv
pip install optimpv

You can also clone your base environment:

conda create -n optimpv --clone base

Additional necessary installs for the agents

Drift-diffusion agent

The drift-diffusion agent uses SIMsalabim to run drift-diffusion simulations.

• SIMsalabim is included as a submodule.
• Install it following the instructions on the SIMsalabim GitHub repository.
• Only works for parallel simulations on Linux. All other optimPV agents work on Windows.

Parallel simulations

To run parallel simulations on Linux you can also install GNU Parallel:

sudo apt update
sudo apt install parallel

Disclaimer

This repository is still under development. If you find any bugs or have any questions, please contact us.