Temperature, Humidity, Ambient Light Monitoring

The goal of this project is to design a temperature and humidity monitoring system for indoor home usage.

Specifications

1. Monitor the temperature (T) in deg celsius with accuracy of better than 1%.

2. Monitor the humidity (RH) in %RH (relative humidity) with accuracy of better than 5%.

3. Monitor the ambient light in relative lux (LUX) with respect to dark.

4. Capable of recording the T, RH, LUX at least once every 30 seconds.

5. Records T, RH, LUX in a database.

6. Presents T, RH, LUX data as a time-series via a web-interface.

7. The web-interface is accesible atleast within the intra-net.

System Design

A block diamgram of the system is shown in the figure below.

Hardware Design

The sensor board has Sensirion SHTC3 temperature and humidity sensor. The board is designed to fit on top or under of the Raspberry Pi Pico W (PicoW).

Software Design

Monitoring Server

The monitoring server is deployed on a Raspberry Pi 400. The sensor node communicates to a Python Django based web-app which can receive data on port 8080. The web app stores data in a SQLite database. The database backup is scheduled for everyday on a internal network attached storage.

Web API Specifications

Web API specifications are documented in Web API Specifications # Web API Specifications

Sensor Node

Send Monitoring Data Point

• Sensor Node sends the data with a <POST> request.

• Endpoint: /thlmonitor/sensor_data

• Send data in as Python dictionary. Individual fields are listed below:

  • location sent as a tuple of grid co-ordinates as per the sensor grid specifications

  • timestamp sent as a of tuple of time.localtime()

  • Temperature T sent as a number with one decimal precision.

  • Relative Humidity RH sent as a an integer number between [0, 100].

  • Ambient Light intensity LUX sent as a number with one decimal precision.

For example: Ground floor, living room

{"location":'0E4',"timestamp":(2023, 1, 15, 16, 47, 39, 6, 15),"T":23.5,"RH":50.5,"LUX":34.2}
OR
{"location":'GE4',"timestamp":(2023, 1, 15, 16, 47, 39, 6, 15),"T":23.5,"RH":50.5,"LUX":34.2}

First floor, N Office room

{"location":'1A6',"timestamp":(2023, 1, 15, 16, 47, 39, 6, 15),"T":23.5,"RH":50.5,"LUX":34.2}
OR
{"location":'FA6',"timestamp":(2023, 1, 15, 16, 47, 39, 6, 15),"T":23.5,"RH":50.5,"LUX":34.2}

Get Current Time from Server

• Sensor requests current time with a <GET> request.

• Endpoint: /thlmonitor/sensor_data

  • Data sent as number of seconds from epoch.

Example Sensor Grid Specifications

User Front End for Home Monitoring

Sensor Node

Programmers model for the sensor board is shown in following figure:

• Each sensor node consists of a PicoW as controller and THL sensor board.

• Each sensor node has a location tag to uniquely identify the location.

• PicoW sets up time of internal RTC clock

• PicoW uses MicroPython with urequest library

• Each of the Sensirion T/RH and LiteOn Ambient light sensor communicates with PicoW via I2C interface.

• The sensor node acquires T, RH and LUX data every 30 seconds and puts it in a queue.

• The node attempts to send the data in queue to the server in JSON format specified above. If successful the datapoint is dequeued, else attempted again.