AI Case Study
St. Vincent's Hospital achieves 20% in energy savings by implementing a predictive energy control system for its HVAC from BuildingIQ
St Vincent's Hospital implemented BuildingIQ's predictive energy consumption software, which models a building's thermal processes. The software then adjusts HVAC settings according to real-time data on weather, occupancy, etc., resulting in a total energy savings of 20% and neat peak energy demand reduction of 10%.
Healthcare Providers And Services
"BuildingIQ formed a strategic partnership with Entro.py and entered into a shared energy savings contract based upon the deployment of BuildingIQ’s Predictive Energy Optimization™ (PEO) software. The first phase of PEO was the establishment of
the historic baseline of energy consumption. The second phase was learning the thermal dynamics of the building. The third phase, optimization, began a few weeks later." According to the vendor its "BuildingIQ system is the only energy management system that predicts energy demand and directly adjusts the HVAC system parameters in real time to optimize energy use. PEO automatically and continuously obtains data on the local weather forecast, the occupancy for the building, energy prices, and tariffs. Based on those inputs, it runs thousands of simulations to arrive at the most efficient HVAC operating strategy for the next 12 hours. BuildingIQ's network operations center maintains oversight of the data for 24/7 anomaly detection, data analysis, and diagnosis to assist on-site facility teams."
From BuildingIQ: "St. Vincent’s owns and operates a large public and private hospital network in Australia with facilities concentrated in Sydney and Melbourne. The firm, Entro.py, which has managed the building management systems (BMS) on the Darlinghurst campus for the past few years, called upon BuildingIQ to help reduce the building’s energy consumption."
"Within weeks of the initial optimization, total energy savings
climbed from 5% to 10%, and continued to climb as the Australian summer season progressed. By the peak month of December, total savings—not just HVAC savings—reached 20%. A total net peak demand reduction of 10% was achieved."
BuildingIQ's "algorithms set about the task of learning the specific dynamic responses of the building to a wide variety of continuously changing conditions— everything from temperature and humidity, to occupancy profiles. It takes some 4-6 weeks for
the model to truly understand the unique dynamics of the building thermodynamically, as people enter and leave, congregate and disperse in various zones, as weather conditions change, and as the train of heating and cooling equipment labors with partial load or hums along at peak efficiency."
Includes data "on the local weather forecast, the occupancy for the building, energy prices, and tariffs" which ultimately feed into the "hundred of parameters that monitor and ultimately control the building’s dynamic responses".