Author: Davide Carecci
Initial commit: 01.05.2026

---

BIOGoAlS.Architecture contains the code requried to run the BIOGoAlS.Twin controller on a Raspberry Pi (RPi; OS is Raspbian ARM 64bit) equipped with (i) GPIO RPi Relay Board, (ii) a peristaltic pump, and (iii) a RJ45 connection between the Awite biogas analyzer and the RPi (comumication via Modbus protocol), as it was done during the real-time control experiment carried out between 03-07/2025 at the BioTA lab (Unversidad Federico Santa Maria, Valparaiso, Chile). For further details about the test facility and equipment, please refer to the author's PhD thesis.

The tool:

• pre-processes measurements from the Awite (biogas flow rate and composition);
• computes a control action running a controller.py;
• sets the ON/OFF duty cycle of the GPIO relay board to activate/deactivate the power supply to the peristaltic pump;
• recursively at each control interval (or even asynchronously when a new biogas composition measurement is available).

---

Documentation

• Python Requirements
  List of Python packages required to run the notebooks and scripts in this repository.

• Example manual
  Instructions to guide the user to correctly run the content of the /Example folder.

---

• pwm_set_R1.py Python code used to read and set the ON/OFF duty cycle of the GPIO RPi Relay Board.

• Other minor custom helper functions.

Inside the /Example folder:

• controller.py: Python code that contains the frame in which the user should call its desired controller code (e.g., the test_closedloop_Chile_main.py file of the BIOGoAlS.Twin repository) to compute the control action. Returns the latter as ON/OFF duty cycle.

• cronjob.py Python code that is actually scheduled to run recursively at each control interval or fixed period of time (e.g., each hour). It calls the functions needed to pre-process data, check for the existence of a new biogas composition measurement and, eventually, trigger controller.py.

• /awite subfolder, that contains:

  • modbus3.py Python code needed to read data from Awite analyzer (via Modbus protocol).
  • preprocess_meas.py Python code that converts the values read via Modbus and stores them in the proper format to be further read by controller.py and cronjob.py.
  • Other minor custom helper functions.

• /Input subfolder.

---

Notes

• !! Note that the data of the biogas flow rate (e.g., measured by Ritter counters) must be transmitted from the counters to the Awite analyzer, before the information is sent to the RPi. You must contact the Awite support to do this step.
• Markdown links (./folder/file) are used for proper GitHub rendering.
• Python 3.10+ is required.
• For further details, please refer also to the PhD thesis of the author (visit POLItesi or request a copy to davide.carecci@polimi.it).
• Run neofetch in terminal of the RPi to see properties.
• The python interpreter and package manager in miniforge3. The base environment automatically activates when starting a terminal. See basic instructions to create a new virtual environemnt called, for example, adcontrol.
• To run any python file (i.e. $filename.py) related to AD control, run in terminal: conda activate adcontrol cd /home/username/NMPC python $filename.py
• You can download a copy of the controller library from https://github.com/DaveCacci/B.Twin.
• To install the required python packages, see /home/username/python_requirements.txt, and to install them, for each $packagename, run in terminal: conda install -n adcontrol $packagename If you want a specific version $packageversion, run instead: conda install -n adcontrol $packagename == $packageversion NOTE You can check which packages are already installed by running: conda activate adcontrol conda list
• Follow the instructions to setup the relay board, creating another folder /home/username/RPI_Relay_Board. A subfolder of the latter, called python, can be used to place the python files that set the PWM of the actuators (e.g., pwm_set_R1.py).

---

© 2026 Davide Carecci — All rights reserved.