Energy Manager

PEAK: Prepaid Energy Assistance Kit

Installation

1. Clone the repository:

git clone https://github.com/HAPPEN-lab/peak.git
cd peak

2. Create and activate a virtual environment:

python -m venv .venv

# Windows
.\.venv\Scripts\activate

# Linux
source .venv/bin/activate

3. Install dependencies:

pip install -r requirements.txt

Usage

Python MILP Models

Navigate to the src directory containing the project's code. Inside the main python file, milp_main.py, choose an optimization algorithm to run. This can be done by setting the constant ALGORITHM to either 'OBM' or 'DFM' at the top of the file. Along with choosing an algorithm, users must input the start day, load_order, data path (for the forecast), column names for loads, and the recharge percentage. Users must additionally add the Gurobi license file if they are running the DFM algorithm. This can be done by uncommenting the following line at the beginning of the script:

# os.environ["GRB_LICENSE_FILE"] = r'/path/to/gurobi.lic'

Once an algorithm is selected and ran, outputs from the model and simulation will be output in the terminal, but results can additionally be accessed by storing the output of the algorithm's main function.

AFG Fractional-Knapsack Model

AFG is designed to be run on an ESP32-S3 with data stored on a connected SD card. It is recommended to compile and run AFG using the platformio.ini file inside this repository. Compilation and flashing to the ESP32 can be done using the following command:

platformio run -e afg -t upload

Additionally, the code can be monitored over the serial port by adding the -t monitor flag. Note: The platformio .ini file contains upload/monitor port options for Linux, MacOS, and Windows. Make sure to choose the port option for your operating system. Also, ensure you are not communicating with another device over serial, or change the port selection to upload/monitor a specific port.