Project overview#

Introduction#

Thevenin is a Python package designed to offer a simple, robust, and flexible interface for running Thevenin equivalent circuit models. Its primary focus is on simulating battery performance under various load conditions, making it an invaluable tool for researchers and scientists working on battery technologies. Whether you need to generate synthetic data, optimize parameters for real-world systems, or integrate fast, accurate models into control algorithms, Thevenin is built to handle it all.

The package provides a balance between simplicity and capability, supporting both constant and dynamic loads, and is scalable enough to fit into real-time systems.

Key features#

  • Flexible Circuit Elements: All circuit elements, including resistors and capacitors, can either be constant or depend on state of charge (SOC) and temperature.

  • Customizable RC Pairs: The model supports any number of RC (resistor-capacitor) pairs, from zero to \(N\), allowing the model to scale from simple to more complex systems.

  • Thermal Modeling: The package can operate in isothermal conditions or simulate thermal effects using a lumped thermal model for greater accuracy.

  • Versatile Experiment Interface: The API allows for intuitive and flexible simulation of any type of load, including constant and dynamic loads driven by current, voltage, and/or power.

  • Cross-platform Support: Written in Python, the package runs on any platform that supports Python and is continuously tested across multiple Python versions.

Use cases#

Thevenin is designed for a variety of applications in the battery research space:

  • Parameter Optimization: The packaged models can integrate with optimization routines, enabling fast model calibration to real-world battery systems.

  • Synthetic Data Generation: Researchers can generate synthetic data for analysis, algorithm development, or system testing.

  • Battery Model Integration: Thevenin is designed for integration with control systems, such as those using Kalman filter algorithms for real-time state estimation and battery management.

Target audience#

Thevenin is built for scientists and researchers in the battery industry. Its primary applications focus on:

  • Battery performance simulation

  • State of health (SOH) estimation

  • Real-time control integration

Users who require accurate, fast models that can be integrated into control algorithms and optimization frameworks will find Thevenin especially valuable.

Technology stack#

  • Language: Python

  • Compatibility: Runs on any hardware that supports Python. Multiple versions are supported.

Project origins#

Thevenin was developed by researchers at the National Renewable Energy Laboratory (NREL) as part of the Rapid Operational Validation Initiative (ROVI), a project funded by the Office of Electricity. The ROVI project aims to streamline the process of validating new battery technologies and chemistries as they enter the market. Thevenin contributes to this effort by providing a tool that models battery performance with flexibility and speed. If interested, you can read more about ROVI here.

Roadmap and future directions#

Thevenin has several exciting long-term goals:

  • Optimization Submodule: A future release will include an optimization submodule for automated parameter fitting to experimental data.

  • Integration with Kalman filters: The package is currently being exercised with moirae, a separate package containing Kalman filter algorithms. This will demonstrate how Thevenin can be used for online state estimation, improving real-time battery management.

Contributions#

The Thevenin project is hosted and actively maintained on GitHub. Developers interested in contributing are encouraged to review the Code structure and Workflow sections for detailed information on the branching strategy, code review process, and how to get involved. All contributions are welcome as we continue to evolve Thevenin’s capabilities.

On this page