Project Overview

-
As our first proper post, we'd like to introduce to you the main concept of our idea, and how we perceive it to work:

An Arduino Nano is located at each water outlet in the home, connected (directly) to a water flow meter. This Arduino is connected via radio to a Raspberry Pi - the hub. Through radio transmissions, the Arduinos will write their data from the flow meter to XML files saved on the hub; the RPi will interpret the data, display it on a local website, and provides tips as to how to save water. This website could be configured (via port forwarding on the router) to be available outside of the home - through either a plain public IP address, or a hostname (using dynamic DNS or a static public IP). We would like the water-saving tips to be powered by an AI system; this is definitely the most challenging part of the system to develop.

For a preview of the locally-served website, please see https://demo.traqua.ml.

All of our code is available open-source on GitHub.

Comments

Post a Comment

Thanks for commenting! We'll get back to you as soon as we can.

Please ensure that your comment isn't offensive or inappropriate.

Popular posts from this blog

Move From XML To CSV

-
Recently, we were discussing our file structure and realised that XML files are unnecessarily complex - we were using the following structure: <livehousewaterusage> <appliance> <appname>sink</appname> <room>kitchen</room> <rate>55</rate> </appliance> <appliance> <appname>sink</appname> <room>bathroom</room> <rate>75</rate> </appliance> <appliance> <appname>toilet</appname> <room>bathroom</room> <rate>85</rate> </appliance> <appliance> <appname>bath</appname> <room>bathroom</room> <rate>80</rate> </appliance> <appliance> <appname>shower</appname> <room>bathroom</room> <rate>100</rate> </appliance> <appliance> <appname>shower</appname> <room>bathroom</room> <rate>
Read more

Website

-
Image
I've finished coding the website! I used the Material Design Lite Dashboard  template ( https://getmdl.io ), and made significant theme and content changes (including creating a dark theme, based on the two main colours in our logo - #1A1A1A & 006DCC). The dashboard now looks like this: It depends on the following resources: Material Design (material.min.js, material-theme.min.css) Material Icons (material-icons.css) jQuery (jquery.min.js) Morris JS (morris.min.js, morris.css) Raphaël (raphael.min.js) Palette JS (palette.js) jQuery CSV is used to retrieve the data from CSV files on the local server, and Morris JS (which depends on Raphaël) is for the charts that are created from the CSV data, and Palette JS is used to create an array of colours forming a rainbow for each doughnut chart. The charts are configured to update every second; the code is all available open-source  on GitHub . We are yet to implement the AI tips service, and are in the process of org
Read more

CSV Reading/Writing

-
We have recently finished coding the Python (used for writing to the CSV files) and jQuery (used to read them), and the code is available on GitHub . However, I include it here for convenience: Outlets: import socket import sys message = raw_input("Enter your message:") # Read message to be sent # Create a UDP socket sock = socket.socket(socket.AF_INET, socket.SOCK_DGRAM) server_address = ('192.168.0.18', 6401) # Server's IP Adress and Port Number for Communication. try: # Send data print (sys.stderr, 'sending "%s"' % message) # Print the message to be sent. sent = sock.sendto(message.encode('utf-8'), server_address) # Send the message # Receive response print (sys.stderr, 'waiting to receive') # Listen for acknowlegement of file receival data, server = sock.recvfrom(4096) # Receive acknowledgement data print (sys.stderr, 'received "%s"' % data) #Print acknowledgement data fi
Read more