How one University Plans to Achieve Zero-Net-Carbon by 2050
- Written by Jim Green
The ability to collect real time data from a variety of building automation systems and metering devices, and to integrate that data to provide detailed, campus-wide energy demand and consumption indicators, has been critical to the University of Minnesota’s energy efficiency success and its continued progress towards carbon neutrality.
The University of Minnesota’s Minneapolis-St. Paul campus is home to over 200 buildings that provide 23 million square feet of teaching, research and support space. Annual energy costs on this sprawling urban campus, located in a northerly region with seasonal extremes of temperature and humidity, exceed $40 million.
Despite the demands of a harsh continental climate, the University of Minnesota has committed to becoming carbon neutral by the middle of the twenty-first century, having signed on to the American College & University Presidents’ Climate Commitment (ACUPCC). Thus, energy efficiency and conservation is a major component of the Climate Action Plan the university has developed to achieve the climate neutrality goal. To that end, an energy management group has sought to save on energy and emissions, taking into account that over 80 percent of the campus’s greenhouse gas emissions originate in steam boilers owned by the university, while the remaining 20 percent is associated with electricity purchased from power plants not owned by the university.
Improvement of campus energy efficiency has resulted in savings of $6 million in recurring annual energy costs since 2008. That success has been the result in large part of the energy group’s ability to automate the collection, display and analysis of real-time energy and demand data, and correlate it with mechanical and electrical system operational data from an extensive, integrated building automation system (BAS). The main focus in that effort has been on heating, ventilation and air conditioning (HVAC) systems, which account for about 70 percent of the university’s energy demand. Hot and humid summers drive demand for electricity-intensive air conditioning, while infamously cold and dry winters make for extensive steam heating provided mainly by natural gas consumption.
In addition to those factors, many of the university’s research buildings require very high ventilation rates in order to ensure the safety of occupants. While a typical office building might be well served by two or three air changes per hour, a research lab may need more than ten or twelve changes an hour.
Thus, although the smart grid often is thought of as a supply-side energy technology, the demand side of the grid—on the back side of the meter—has been where the action is as the university works to improve efficiency.
In order to allow for competitive bidding, the university has not specified that any one manufacturer’s building automation system be installed in campus buildings. The university has insisted, however, that all BAS communicate via the internationally accepted BACnet protocol. This allows for single–seat operation and monitoring of all buildings on campus. At the same time, any university employee with the proper credentials is able to log into the firewall-protected system from any Internet connection to monitor building conditions and system operation. System integration with open access to information has resulted in several successful energy efficiency initiatives.
There is no easier way to save energy than to turn off (or slow down) energy consuming equipment. Many of the air handling units (AHUs) that consume so much energy have fans with motors exceeding 200 horsepower. With modern direct-digital-control building automation systems it is a relatively simple task to schedule fans, pumps, chillers and other energy intensive components to run according to time-of-day schedules.
Many campus buildings have well defined business hours, allowing for efficient scheduling of equipment to minimize energy consumption and emissions. Scheduling is done centrally by operators who are able to access all of the operating schedules from a central BAS center. System operating schedules are adjusted weekly or as often as required based on class schedules or special events.
In a perfect world, there would be no more to the story. But in the real world, systems malfunction, systems are switched from automatic operation to manual operation for troubleshooting or issue resolution (and not switched back to automatic operation), or scheduling errors are made. In order to detect these sorts of issues the university has leveraged the integrated BAS systems to create a fan schedule variance monitoring process which compares actual weekly cumulative fan operating hours to the scheduled weekly operating hours.
Fans running more hours than scheduled are potentially wasting energy; fans running fewer hours than scheduled may indicate a maintenance issue or system failure. In either case, the data is centrally reviewed, problems flagged and investigations initiated.
The university has documented over $750,000 in annual energy cost savings with this simple strategy made possible by the integration and networking of individual BAS systems. Taking the concept one step further, facilities engineering staff have developed pre-programmed demand reduction routines that can be implemented campus-wide with a few strokes on a keyboard and click of a mouse. While initially developed to mitigate risk during periods when energy supplies are curtailed or limited by a system failure, the potential exists to refine the routines for use as everyday campus-wide demand reduction strategies.
Knowing well the old adage that you can’t manage what you don’t measure, the university meters each type of energy (electricity, steam, chilled water, natural gas) at each building. Live meter data is collected, processed and made available for real-time, building-by-building “dashboard”-type display and statistical analysis. Just as air handler fans are “caught misbehaving” by the schedule variance monitoring process, abnormal building energy consumption data can be detected, investigated and resolved in near real time. Rather than being surprised by then monthly utility bill weeks after the fact, meter readings are centrally analyzed and abnormal data are flagged and investigated daily.