Buildings Magazine - August

The recent history of building opera- tions management has been defined by the steady improvement of the tools that operators and engineers can use to inform decision-making. Building information modeling and other analytical models, whether data or physics-based, have had significant positive impacts on the industry.

Nevertheless, the American Council for an Energy-Efficient Economy found that most commercial air conditioners in the U.S. are oversized by 25-50%.

The progress hasn’t been sufficient. A shift is needed from the “art” of intuition, assumptions and rules of thumb to the “ science” of empirical evidence driving statistically significant conclusions.

The Internet of Things (Io T) has unlocked the potential to collect real-time empirical data about the individual components that make up buildings. With data, analytics software can focus on benchmarking, comparisons, and identifying patterns and anomalies from observed performance.

Measuring and analyzing actual performance is necessary to continue to eliminate waste in capital investments, maintenance and repairs, and energy consumption.

Even robust building management systems (BMS) that use continuous data inputs to directly control equipment usage are prone to waste through intuition and assumptions. Improperly configured BMS are believed by the Office of Energy Efficiency & Renewable Energy to account for 20% of building energy usage (about 8% of total energy usage in the U.S.).

Let’s follow the lives of two commercial air conditioning units.

Oversized Air
Conditioner Fails

The first one was installed and maintained according to industry best practices. When designing the system, the engineers used rule of thumb calculations to ensure that it guarantees enough cooling capacity to satisfy tenant requirements. As is common, the system is 25% oversized.

The preventative maintenance schedule, based on the manufacturer’s recommendations, dictates that the unit be serviced once per year in early spring before the cooling season.

After installation, tenants grumble that the indoor temperature varies wildly when it’s hot out. They find that when the air conditioner kicks in, it gets very cold quickly and then gets warm before jumping down again. They also notice that it seems to stay humid even when the air is cold.

Eventually, the operators get a complaint from a tenant during the summer that its space isn’t cooling at all. After an investigation, operators uncover that the unit has been short cycling. They check the levels of refrigerant, test the thermostat to ensure it’s reading correctly and is appropriately placed, and make sure the low-pressure control switch and compressor are working properly.

Everything appears to check out, so the operators determine the unit is likely oversized and is cooling the space too quickly, cycling on and off quickly to maintain the desired temperature. While the solution is to replace the unit, there’s no room in the capital expense budget, so the operations team decides there’s nothing they can do except increase the amount of preventative maintenance checks during the cooling season.

For the rest of its life, the unit wastes energy by running when the building is unoccupied and outside air temperature is relatively low. After 15 years, the unit fails entirely, and a replacement is required.

Benchmarking for the
Correct Unit

Now let’s look at the life of the second air conditioning unit. It was installed and maintained using empirical data from circuit-level electrical demand sensors.

When designing the system, the engineers used equipment-level benchmarking to determine the right unit by considering factors such as make and model, climate, sizing and occupancy schedule. The benchmark was built by tracking and recording millions of machine hours of air conditioning units across the portfolio and aggregating data from other equipment in the same region and building vertical.

As the manufacturer recommends, the
operators perform a preventative mainte-
nance check in the early spring before the
cooling season. In addition, the operators
receive notifications when the electrical
demand of the unit indicates that conditions
necessitate servicing the unit. When the unit
inevitably needs repair, a fault detection
alert directs operators to the unit, avoiding
an investigation and enabling them to fix
the system before tenants notice.

If at any point, the system configurations are changed to a suboptimal schedule or balance point, the operators receive a notification that uses inferred occupancy and weather data to prescribe the optimal settings. The unit lasts its full 20-year lifetime before a replacement is required.

What’s the Difference?

The second scenario results in reduced costs and improved tenant comfort. It also requires continuously tracking electrical consumption at the circuit level. This is an investment that that pays for itself when all the avoided costs are added up:

■ Capital investment costs: Lowered by right-sizing equipment rather than paying for unnecessary capacity.

■ Maintenance costs: Lowered by reducing the number of unplanned maintenance work orders, eliminating the time spent investigating tenant complaints and avoiding arbitrary additional maintenance checks.

■ Energy costs: Lowered by maintaining
the optimal schedule and set points
instead of letting performance drift.

■ Net present value of money: Because future money is less valuable than money today, value is added by delaying equipment replacement for five years.

■ Tenant experience: Improved by ensuring normal air quality and by proactively addressing issues, potentially affecting leasing decisions.

The cost of an Io T-based equipment energy tracking solution is generally an order of magnitude lower than a traditional building management system. Although these systems can’t control equipment like a BMS, the data-driven decision making enabled by these solutions present an attractive investment, whether there's a BMS in place or not.

Experienced operators and engineers will always be necessary to ensure that systems run properly, but should favor hard data over assumptions. The more we can direct building operations from an art to a science, the healthier our indoor environment will be, the easier operators’ jobs will be and the more profitable real estate will become.