Sydney, Australia and Valencia, Spain --- (METERING.COM) --- May 31, 2012 - The development of energy saving solutions is proving to be a popular task for university students around the globe.
At the University of Sydney's Faculty of Engineering and Information Technologies, two PhD students, Mahboobeh Mogaddham and Waiho Wong, have developed the “MyPower Energy Platform” to monitor the power use of individual appliances and help consumers decide how and when to use or replace them.
And at the Universidad CEU Cardenal Herrera in Valencia, students in the Computer Engineering in Information Systems department are working with researchers on the application of neural networks in homes to “learn” the energy consumption behaviors of the inhabitants and implement energy saving actions.
The MyPower Energy Platform comprises a smart plug with an embedded GSM unit that monitors major household appliances, such as washing machines, clothes dryers, microwaves, electric water heaters and refrigerators, and transmits the information via SMS to a cloud-based data warehouse every 30 minutes. Via the MyPower website, the householder can then access their consumption data and drill down to the individual appliances' costs, based on the prevailing peak, shoulder or off-peak rates. The householder is also able to remotely schedule or switch off the appliance via the smart plug.
"The solution captures highly disaggregated data at the appliance level and transforms it into actionable knowledge via analytics-based applications," explains Professor Joseph Davis, director of the Knowledge Discovery and Management Research Group, who is supervising the two students.
According to the students, householders should be able to reduce their consumption by up to 10 percent using the platform.
The basis of the Valencia students and researchers’ solution is a powerful database of data on the temperature, humidity, lighting, air conditioning and other energy consumption of a home’s inhabitants, as well as on the production of energy from home generation sources. From this the consumption patterns can be established and optimal relationships established between these and the generation patterns, from which recommendations can be made on the optimum times to switch on appliances or for example run the air conditioning.
The solution uses artificial intelligence algorithms, and apparently from ten days to a month of data is required to “learn” routines and efficiently manage a home from the energy standpoint. Savings of at least 5 percent should be obtained.
This project is being managed by Nicholas Montes, coordinator of computer engineering at the CEU-UCH.