RA: Smart Plug-based Appliance Energy Profiling & Prediction Portal for Green Buildings

Research Assistantship Available

Smart Plug-based Appliance Energy Profiling and

Prediction Portal for Green Buildings

New emerging “smart plugs” embed a micro-controller and low-power communication device that allows you to monitor the power consumption of individual devices (e.g., microwave, coffee machine, laptop) plugged into power sockets, and communicate such power consumption information over a wireless network to a central monitoring station. Such devices could lead to substantial savings of energy and money by enabling Internet-based monitoring and real-time control of the behavior of individual appliances. This project will use real-life microcontroller kits (ACME Plugs from Moteware) and real-life building measurement data to explore whether such measurement-based monitoring can be used to:

• Develop Smart Circuit Breaker — i.e., to lessen the burden of the user of plugging each and every appliance/device in the building with a smart plug; we will investigate connecting multiple devices together with an individual smart plug/smart circuit breaker and design a smart circuit breaker using energy metering chip (ADE7753), AC/DC power supply, Microcontroller with radio (TI MSP430F16 and CC2420 radio supported by TinyOS) and solid state AC relay (Sharp S216SE1) etc.
• Profile individual devices — i.e., use NILM (non-intrusive load monitoring) data analytics algorithm on the time-series of power consumption traces to infer the type of plugged-in device (e.g., distinguish between a laptop & a coffeemaker), thereby building a dynamic catalog of the types & number of devices connected by a consumer.
• Predict the power consumption of individual rooms — i.e., using the past history of the power consumption of individual devices to create predictive inferences of the usage patterns for individual devices (e.g., learn that the individual switches on a dehumidifier for ~3 hrs every Thursday).

Expertise: Technical knowledge of standard time-series & statistical mining techniques (e.g., regression, support vector machines) is needed. Significant programming knowledge of Java & ability to create simple Web Applications is a must. Knowledge of TinyOS, Embedded System and Networking protocols is a plus, although not essential. The project will utilize real ACME plugs, which are programmed using TinyOS & which communicate using a ZigBee radio.

Please contact Dr. Nirmalya Roy at Sorry, you need javascript to view this email address. for research assistantship for this project.

• More Information about RA: Smart Plug-based Appliance Energy Profiling & Prediction Portal for Green Buildings