Implementation of an Adaptive Occupancy and Building Learning Temperature Setback Algorithm (OR-16-015)

The occupancy patterns in office buildings have been becoming increasingly diverse. And, even in cases where the occupancy periods are still rigid, the time needed to bring a room from nighttime setback temperatures to the setpoint temperatures not only change in time but also vary between offices. Consequently, operators have been challenged to choose conservatively short temperature setback periods. In recognition of these challenges, a self-adaptive control algorithm that can learn the recurring occupancy patterns and the parameters of a model predicting the indoor temperature response was implemented in a southwest-facing shared office space in Ottawa, Canada.