Dynamic clustering of temporally incremental energy consumption patterns in a knowledge cloud

This project will develop a new mechanism for grouping objects in a dynamic environment, where new objects are regularly added with limited or incomplete information. Furthermore, the information about the existing and new objects increases over time. This new grouping mechanism will be called dynamic clustering of temporally incremental patterns. The proposal will be tested using energy consumption patterns for a large number of buildings. The types of the buildings will vary based on their usage such as office buildings, warehouse, shopping malls, hospitals, educational institutes, etc. The buildings will further differ from each other in terms of size, occupancy, and hours of operations. These buildings will also be situated in geographically diverse locations and climatic conditions. The data stored in the cloud will be analysed using a number of statistical and artificial intelligence techniques to create a knowledge repository. The knowledge will provide the system abilities to:
predict energy consumption for any given day depending on date, time, and weather conditions for any one of the buildings stored in the cloud
optimize the energy management system to provide comfortable operating conditions with minimum energy consumption
accommodate new buildings without knowing their history of energy management

Faculty Supervisor:

Pawan Lingras

Student:

Ilia Pavlovski

Partner:

Green Power Labs Inc.

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Saint Mary's University