Developing a new workflow for occupancy estimation using Wi-Fi sensing for building energy simulation

The current study aims to develop a new workflow for occupancy schedules estimation using WiFi sensing technology and utilizing different machine learning and deep learning algorithms. Aerial will provide preprocessed CSI data, occupied and non-occupied time series, and activity level time series to be used by intern for extracting occupancy presence pattern, the number of people estimation and pattern extraction, and human activity estimation. Pattern extraction and analysis of the three mentioned datasets will provide a basis for estimating four occupancy presence, occupancy activity, lighting, and electrical equipment usage schedules. The estimated occupancy schedules will be fed as an input of EnergyPlus (a building energy simulation software) along with other energy-related data to calculate the building heating and cooling demand. In the end, a short report will be provided to occupants as feedback of their behavior by analyzing the energy demand and occupancy behavior.

Faculty Supervisor:

Ursula Eicker;Nizar Bouguila

Student:

Partner:

Aerial Technologies Inc

Discipline:

Engineering

Sector:

Professional, scientific and technical services

University:

Concordia University

Program: