PLR learns and extrapolates the correlation between device settings and building space environments under a variety of conditions, including internal and external temperature and humidity, space occupancy, time of day, and fixed BMS settings. When fully trained, PLR determines what device settings will be optimal under anticipated conditions, as determined by the history of the building, regularly-updated weather forecasts and load schedules. PLR correlates device settings with the comfort parameters of the space.
For example, PLR is able to calculate and remember the degree of degradation of space air temperature, humidity and other space air quality factors when the velocity of a variable speed drive fan is decreased from 80% of its design specification to 60% for 15 minutes during a period of two hours when a room is scheduled to be full.
A reduction in energy consumption can be achieved without causing noticeable discomfort and, in some scenarios, can actually increase human comfort by reducing temperature drift and deploying a comprehensive daily energy consumption strategy. PLR accesses a menu of available actions – intelligent agents – and deploys one or more of them to change device settings in incremental amounts for limited periods of time. This process continues throughout the day, every day, fine tuning device settings in real time and—in the process–reducing energy consumption by measurable amounts.