Intelligent Smart Switch using IoT and Machine Learning

Home Automation Systems have increased worldwide popularity in the last decade or so. But now world looking towards Intelligent Smart Home Automation Systems which must be wise to provide efficient electricity usage plan. Intelligent Smart Switch is based on making user life easier. The Internet of

Project Title

Intelligent Smart Switch using IoT and Machine Learning

Project Area of Specialization

Internet of Things

Project Summary

Home Automation Systems have increased worldwide popularity in the last decade or so. But now world looking towards Intelligent Smart Home Automation Systems which must be wise to provide efficient electricity usage plan. Intelligent Smart Switch is based on making user life easier. The Internet of Things makes a vast amount of data. Additionally, it contains a huge number of sensors and their data which can control or monitor appliances. This project shows about how the Internet of Things (IoT), Machine Learning and Data Mining use for converting the normal home to the Intelligent Smart home. A user of this system can control their appliances using their smartphone application as well as the web browser which are connected to the internet. The user will also get efficient usage plan to minimize the power wastage. This project is divided into four section, Hardware, User Interface (UI), Cloud Platform and Data mining & Machine Learning.

Project Objectives

As an increase in population and global integration, energy consumption also increases. This pretense a risk for run-down natural resources, and Increases demand of renewable energy systems. The prices of fossil fuel going up progressively. Pakistan is in front of the energy crisis where electricity generation is far less than consumption and 50 percent of electricity is consumed by household in Pakistan. By using intelligent smart switches we will be able to reduce energy wastage by providing efficient usage plan. It does not lie that the world is moving towards energy crises. The need of the hour is to save energy by using modern technologies which involve Internet of Thing (IoT) and Machine Learning. The problem is to monitor energy usage and provide an effective plan that helps the user in both energy saving and reducing electricity bills.

The aim of the project is to eliminate energy wastage by monitoring energy usage of each appliance in the home and provide an energy proficient plan to the user. It will be able to predict the behavior of appliances operating in the location and provide a competent profile of the user. Our intention is to use modern technologies and advance knowledge to capitalize on the quality measures.

1. Fabrication of smart switch.
2. Design and development of User Interface (Android app, Web page).
3. Data collection on cloud platform.
4. Effective energy usage plan using Machine Learning.

We can optimize various parameter using Machine Learning such as Bill, Load and time. We want to achieve effective usage plan using different machine learning algorithms.

Project Implementation Method

There are two ways to implement this project.

The first way is to implement by replacing the conventional switch and appliances can only be controlled by  Andriod application using a smartphone or by Web app using laptop or desktop computers.

The second way is to connect these switches parallel to the conventional switch and the appliances can be controlled by both conventional switches as well as Android and web application

We only need wifi to connect to implement this project in any area.

Benefits of the Project

Control your devices from anywhere in the globe.

Adds Safety Through Appliance.

Trim Your Energy Costs.

Increases Convenience Through Temperature Adjustment.

Saves Time and Cost Effectiveness.

Keep eyes on which device is wasting energy.

Ease to control appliance for older and disable persons.

Technical Details of Final Deliverable

This project is divided into Hardware, User interface, Data collection, and Machine Learning algorithms.
Hardware is based around Wemos D1 mini Esp8266 microcontrollers. AC appliances are switches using the relay which is isolated using optocoupler and control by a latch circuit. Current is measured by ZMCT103C Current Transformer by following formulas:
 VPP = getVPP();
 CurrThruResistorPP2 = VPP2 / 380.0;
 CurrThruResistorRMS2 = CurrThruResistorPP2 * 0.707;
 CurrentThruWire2 = CurrThruResistorRMS2 * 1000;
The calculated current is upload to thingspeaks to a channel specified by write key of the channel. Current data is stored upto three decimal places.
We can download this data in the form of an excel sheet as well as XML form. 
Android application is developed by using the android studio in java language. Application has a login page and some other pages including buttons to switch relays. When the status of a button changes from off to on it changes the in Real-time database of Google Firebase from 0 to 1. Then the microcontroller read the tag from the google firebase and changes the microcontroller pin from low to high.
The web application is developed using HTML language and has a similar procedure as an android application to control appliances.

The other sensors also send data to thing speaks. The other sensor includes occupancy sensor, temperature sensor, gas sensor, flame sensor, soil sensor, and rain sensor.
Various Machine learning algorithms applied to the data collected on thing speaks platform to get proficient and efficient energy consumption plan.

Final Deliverable of the Project

HW/SW integrated system

Type of Industry

IT , Energy

Technologies

Artificial Intelligence(AI), Internet of Things (IoT), Big Data

Sustainable Development Goals

Affordable and Clean Energy, Responsible Consumption and Production

Required Resources