Data Science & Communication 2024 Group 2

Topic: Electricity Monitoring and Coordination
Repo: https://github.com/G36maid/electricity_monitor

---

For Teachers

The slides are included in this repository as well as how the work was divided

---

Team Members

Name	English Name	University	Major	Year
羅願群	Jethro	NTNU	Computer Science	1
鍾詠傑	Jie Chung	NTNU	Mechatronic Engineering	2
名嘉山 栞	Shiori Nakayama	KU	ISI	4
ローイ ベンジャミン	Benjamin Joshua Lowy	KU	ISI	2

---

Project Ideas

• Monitor power consumption and power generation:
  • Use data from monitored power → map power consumption → Serve power during emergencies or during vulnerable, high usage times.
  • Power usage on personal devices (app) ← connected to a smart meter.
  • Use algorithms like max flow to visualize the electrical load in each city.
  • Combine with weather data to predict how electricity use varies with weather and season.

usage

This project includes a Linux-based electricity monitoring system using MQTT to send and receive data. It consists of two main components:

1. Linux Client: Monitors power consumption and sends data to the server.
2. Server: Receives data from clients and logs it into a CSV file.

Prerequisites

• Python 3: Ensure Python 3 is installed.
• MQTT Broker: An MQTT broker must be running and accessible. (e.g., 140.122.185.98 on port 1883)
• Dependencies: Install the necessary Python library and tools:
  • paho-mqtt
  • powerstat (for Linux clients)

Installation

For Linux Client

1. Install Python Dependencies:

   pip install paho-mqtt

2. Install powerstat:

   sudo apt-get install powerstat

3. Client Configuration:

   • The client script will prompt for Latitude, Longitude, Building ID, and Category.
   • Make sure to modify the MQTT credentials and broker address as needed.

For Server

1. Install Python Dependencies:

   pip install paho-mqtt

2. Server Configuration:

   • The server script listens for incoming data and logs it into data/electricity_consumption.csv.
   • Ensure the server script connects to the correct MQTT broker and topic.

Running the System

Start the Server

1. Open a terminal and navigate to the directory with the server script.
2. Run the server script:

   python server.py

Start the Client

1. Open a terminal and navigate to the directory with the client script.
2. Run the client script:

   python linux_client.py

3. Enter the required information when prompted.

Notes

• Make sure the MQTT broker is accessible and properly configured.
• Adjust the MQTT broker details and credentials in the scripts as needed.
• The powerstat tool is Linux-specific; alternative methods are needed for macOS or Windows.

---

This README.md provides a clear overview of setting up and running your electricity monitoring system without including the actual code.

Potential Data Sources

• Taiwan Power Open Data:
  • Electricity consumption data.
  • Electricity supply data.
• Additional Datasets:
  • Residential Power Usage 3 Years Data - Timeseries
  • Indian Cities Electricity Consumption 2017-19
  • Energy Consumption of the Netherlands

---

Team Skills and Specializations

• Jie Chung:
  • Network and system administrator.
  • Docker and Kubernetes deployment.
  • C, C++, Rust (not strong in Python).
• Jethro:
  • Data preprocessing, familiar with multiple programming languages.
• Shiori:
  • Data science (visualization), Python.

Summary of Group Chat Day 4

1. Project Focus:

• The group agreed to focus on electricity monitoring and coordination.
• The concept involves using power consumption and generation data to create visualizations that can aid in managing electricity, especially during emergencies or high usage times.

2. Data Strategy:

   • The group decided to simulate fake data for the project.
   • This data will be used for visualizing power consumption and generation on a smaller scale (e.g., a building or district) rather than an entire city, due to the complexity of generating realistic large-scale data.

3. Visualization Tools:

   • Grafana and QGIS were discussed as potential tools for visualization.
   • QGIS was favored, and the group will explore using it, particularly with data that includes geographic coordinates (longitude and latitude) and time series.

4. Technical Considerations:

   • Data will need to include specific power units (GW, MW, kW, W) and possibly be aligned with weather data to predict usage patterns.
   • The group acknowledged the challenge of managing large datasets, especially when simulating data for numerous buildings.

5. GitHub Repository Setup:

   • A GitHub repository was set up for the project: https://github.com/G36maid/electricity_monitor.
   • All group members shared their GitHub accounts and agreed to use Git for code management and collaboration.

Next Steps:

• Data Preparation: The group will focus on generating or collecting suitable data, preparing it for use in visualizations, and ensuring it can be integrated into QGIS.
• Coding and Visualization: Python code will be written to prepare for the demo, and the group will work on integrating the data into QGIS for visualization.
• Collaboration: The team will continue using GitHub for collaboration and code management.

Time Remaining:

• Approximately 20 minutes were left for the immediate tasks at the time of this discussion.