Air Quality Monitoring Kit

The software component of the kit is a web server based on FastAPI Python package for displaying in real time the Air Quality Index (AQI), as well as the particle concentration for different particle diameters. Other air pollutants (Total Volatile Organic Compounds, Nitrate Oxides, Carbon Dioxide, etc) and environmental parameters (Temperature, Pressure, Relative Humidity, Noise Level, etc) are also shown. The data is read from different sensors and stored into a local database before being delivered over websockets periodically to the web page shown in a web browser.

The hardware component relies on a Raspberry Pi as the main processing unit and is designed such that sensors can be easily added, replaced or removed.

Software Prerequisites

• Debian 12

• An InfluxDB 3 Core database is used to store the data received from different sensors and retrieved it to be shown in a web page. The following databases must be configured: dust, gas, climate, sound, light. A read/write access token must be configured.

• Installing lgpio Python module requires to install from sources the official lgpio C library from its GitHub repository

  wget https://github.com/joan2937/lg/archive/master.zip
  unzip master.zip
  cd lg-master
  make
  sudo make install

• A Python virtual environment must be setup in order to run the Python scripts:

    python3 -m venv .venv
    source .venv/bin/activate
    pip install -r requirements.txt

• The Python module for ZMOD4510 O3 NO2 sensor must be added separately

  cd renesas_zmod4510
  pip install .

Setup Influxdb3 Database

• Show database service status

sudo systemctl status influxdb3-core

• Show database service logs

sudo journalctl -fu influxdb3-core

• Quick fix in case the service does not start

sudo rm -rf ~/.influxdb/data/airquality/wal
sudo rm -f ~/.influxdb/data/airquality/snapshots/*

• Create database

  export INFLUXDB3_AUTH_TOKEN="<token>"
  influxdb3 create database <name>

• Show available databases:

  export INFLUXDB3_AUTH_TOKEN="<token>"
  influxdb3 show databases

Obtain Sensor Data

Several Python scripts must be started to read data from sensors and write the data into the database:

• dust_sensor.py: Reads the particle concentration data from two PMSA003 air quality dust sensors, computes the 10 min AQI and writes it to the aqi and pm databases. The USB ports where the air quality sensors are attached must be provided as inputs.

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./dust_sensor.py /dev/ttyUSB1 /dev/ttyUSB2

• noise_sensor.py: Reads the ambiental noise level from a 2-Mic array sensor and writes it to the database.

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./noise_sensor.py

• ambient_sensor.py: Reads the temperature, the relative humididy, the pressure, the gas resistance and the indoor air quality and writes it to the database. In order to read this sensor and process raw data with BSEC library the user must install this Python extension for BME68x.

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./ambient_sensor.py

• light_sensor.py: Reads the visible light in lux and the UV index from an LTR390 sensor.

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./light_sensor.py

• carbon_dioxide_sensor.py: Reads the CO2 level in ppm from an SCD41 sensor.

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./carbon_dioxide_sensor.py

• o3_no2_sensor.py: Reads the O3 and NO2 concentrations in ppb from a ZMOD4510 sensor.

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./o3_no2_sensor.py

• co_sensor.py: Reads the CO concentration in ppm from a ZE07 CO sensor.

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./co_sensor.py /dev/ttyACM0

The scripts print in the standard output the current data read from the sensors and can be installed as services using the services/manage_services.sh script. When installing the Python scripts as services, one must provide in a separate file /etc/default/aq_dashboard.env the database access token. Also, in order to automatically restart the services if an error occurs, the user running the services must have rights to run "sudo systemctl restart *.service" without requiring a password. The datasets provided by these scripts can be analyzed with the aq_data_analysis project.

Web Server User Interface

Start the server with:

    export INFLUXDB3_AUTH_TOKEN="<token>"
    ./aq_dashboard.py

Access the server at: https://<server URL>:8888

Left Column Values (Air Pollutants):

• Nitrate Dioxide (NO2) concentration in PPB

• Ozone (O3) concentration in PPB

• Carbon Monoxide (CO) concentration in PPM

• Total Volatile Organic Compounds (TVOC) index

• Nitrate Oxides (NOx) index

• Carbon Dioxide (CO2) concentration in PPM

• Radon gas concentration in Bq/m3

Center Display:

• The timestamp expressed in local time zone of the most recent dust particle measurement

• The 10-minute Air Quality Index (AQI) - derived from dust particle measurements using a pair of PMSA007 sensors

Right Column Values (Environment Parameters):

• Current room temperature in Celsius degrees

• Relative humidity (amount of moisture in the air) - the normal value is around 40%

• Atmospheric pressure expressed in hectoPascals

• Noise level in the environment in decibels (30 dB would be very quiet)

• Visible light expressed in Lux

• UV index

Particulate Matter Measurements (from PMSA007 sensors):

• PM1.0: Concentration of tiny particles smaller than 1.0 microns in diameter

• PM2.5: Concentration of fine particles smaller than 2.5 microns - can penetrate deep into lungs. These values are used to compute the 10-min. AQI.

• PM10: Concentration of particles smaller than 10 microns - includes dust, pollen, and mold

Particle Count Measurements (from PMSA007 sensors):

• PM0.3+(0.1L): Number of particles larger than 0.3 microns per 0.1 liter of air

• PM0.5+(0.1L) Number of particles larger than 0.5 microns per 0.1 liter of air

• PM1.0+(0.1L) Number of particles larger than 1.0 microns per 0.1 liter of air

• PM2.5+(0.1L): Number of particles larger than 2.5 microns per 0.1 liter of air

• PM5.0+(0.1L) Number of particles larger than 5.0 microns per 0.1 liter of air

• PM10+(0.1L) Number of particles larger than 10 microns per 0.1 liter of air

The notification list shown in the right image is sorted by timestamp in descending order and can be accessed by clicking the bell icon.

Bill of Materials

Type	Quantity	Comment
Processing Unit	1	Raspberry Pi 5, 4 GB RAM, 128 GB SSD
Dust Sensor	2	Plantower PMSA003 UART
Noise Sensor	1	Seeed ReSpeaker Lite Kit-USB 2 Mic Array
Temperature/Humidity/Pressure/Gas Sensor	1	Bosh BME688 I2C
Light Sensor	1	Lite-On LTR390 I2C
CO2 Sensor	1	Sensirion SCD41 I2C
TVOC and NOx Sensor	1	Sensirion SGP41 I2C
O3 and NO2 Sensor	1	Renesas ZMOD4510 I2C
CO Sensor	1	Winsen ZE07 CO UART
Radon Sensor	1	Airthings Corentium Home 2 BLE

Reference

https://cristeab.medium.com/building-an-air-quality-monitoring-station-ba74098f0528