eta-utility 3.2.3

A framework for researching energy optimization of factory operations

While there are many tools which are useful in the area of energy optimized factory operations, at the ETA-Fabrik at Technical University of Darmstadt we have recognized a lack of comprehensive frameworks which combine functionality for optimization, simulation and communication with devices in the factory.

Therefore, we developed the eta_utility framework, which provides a standardized interface for the development of digital twins of factories or machines in a factory. The framework is based on the Gymnasium environment and follows a rolling horizon optimization approach. It provides standardized connectors for multiple communication protocols, including OPC UA and Modbus TCP. These facilities can be utilized to easily implement rolling horizon optimizations for factory systems and to directly control devices in the factory with the optimization results.

Full Documentation can be found on the Documentation Page.

You can find the source code on github. If you would like to contribute, please check our working repository.

Warning

This is beta software. APIs and functionality might change without prior notice. Please fix the version you are using in your requirements to ensure your software will not be broken by changes in eta_utility.

The package eta_utility consists of five main modules and some additional functionality:

• eta_x is the rolling horizon optimization module which combines the functionality of the other modules. It is based on the gymnasium framework and utilizes algorithms and functions from the stable_baselines3 package. eta_x also contains extended base classes for environments and additional agents (or algorithms).

• The connectors module provides a standardized way to connect to machines and devices in a factory or other factory systems (such as energy management systems). The connectors can also handle subscriptions, for example to regularly store values in a database.

• The servers module can be used to easily instantiate servers, for example to publish optimization results.

• simulators are interfaces based on the fmpy package which provide a way to simulate FMU (Functional Mockup Unit) models. The simulators can be used to perform quick complete simulations or to step through simulation models, as would be the case in rolling horizons optimization.

• timeseries is an interface based on the pandas package to load and manipulate timeseries data from CSV files. It can for example rename columns, resample data in more complex ways such as multiple different resampling intervals or select random time slices from data. The scenario_from_csv function combines much of this functionality.

• Other functionality includes some general utilities which are available on the top level of the package.

Some particularities

If you want to have logging output from eta utility, call:

from eta_utility import get_logger
get_logger()

eta_utility uses dataframes to pass timeseries data and the dataframes are ensured to contain timezone information where sensible.

Citing this project

Please cite this project using our publication:

Grosch, B., Ranzau, H., Dietrich, B., Kohne, T., Fuhrländer-Völker, D., Sossenheimer, J., Lindner, M., Weigold, M.
A framework for researching energy optimization of factory operations.
Energy Inform 5 (Suppl 1), 29 (2022). https://doi.org/10.1186/s42162-022-00207-6

We would like to thank the many contributors who developed functionality for the package, helped with documentation or provided insights which helped to create the framework architecture.

• Niklas Panten for the first implementation of the rolling horizon optimization now available in eta_x,

• Nina Strobel for the first implementation of the connectors,

• Thomas Weber for contributions to the rolling horizon optimization with MPC algorithms,

• Guilherme Fernandes, Tobias Koch, Tobias Lademann, Saahil Nayyer, Magdalena Patyna, Jerome Stock,

• and all others who made small and large contributions.

Contributions

If you would like to contribute, please create an issue in the repository to discuss you suggestions. Once the general idea has been agreed upon, you can create a merge request from the issue and implement your changes there.