Abstract:
This thesis presents a new decision support approach to energy control and monitoring
system of domestic appliances. In the modem world, people are rapidly turning to
technology as a fast and cost-effective way of improving quality of daily living. This
primary goal is to address the needs of the end user by employing networked low-power
sensors sensitive to the environment, so it can be altered to their liking.
The proposed system consists of following steps: energy control and monitor, data
analysis and data predictions. This research will present the design and implementation of
a practical and simple smart home system, which can be further extended. The system is
based on: group of sensors, Arduino UNO with unit and WIFI as a communication
protocol.
These devices can be easily controlled via user-friendly interfaces via web applications.
The web applications are available for Consumers and Administrative Staff. Those web
applications represent to the users are statistical data by using Google charts.
Data analysis part has done using Data Mining techniques such as clustering and
regression analysis. Sample data has been generated by using Test Data Generation Tool
is DTM tool. Clustering and Regression Analysis has been done by using Rapid Miner
Tool. Data prediction was done by using Regression Analysis technique.
The main advantage of the proposed system is that it is a sensible, secure and easily
configurable system that provides end users with a cost-effective energy consumption
solution.