The process of commissioning buildings and optimizing utility systems is evolving along with utility information technologies.
This article presents a detailed analysis of a project carried out at the David Skaggs Research Center in Boulder to identify operational issues, opportunities for increasing the efficiency of utility systems, and other improvements after the start of building operation in the first year after completion.
The optimization is expected to pay for itself in less than a year. It improves the quality of the utility services provided, allowing the building to meet the needs of customers more reliably and efficiently.
While the utility improvement process was considered in the context of new construction in this case, it can also be used for the renovation and recommissioning of existing utility systems, for which a relatively quick payback has been demonstrated.
The findings also indicate that further advances in utility information technology will make these diagnostic techniques and their associated benefits more widely available.
This will fundamentally change the process of commissioning buildings and their subsequent use.
Project History
The U.S. General Service Administration (GSA) created the David Skaggs Research Center to serve the National Oceanic and Atmospheric Administration's (NOAA) Environmental Research Laboratory, the National Weather Service, and the National Geophysical Data Center, among other federal agencies. The complex, where the research center is located, contains 720 offices with operable windows, 20 conference rooms, several computer centers, and nearly 100 wet and dry technology labs. The complex consists of three floors above ground and one underground. The total area of the center is 34,500 square meters.
To provide heating for the complex and a year-round supply of hot water, two gas boilers are used, with an efficiency of 82% in steady-state operation. The capacity of each boiler, installed at an altitude of 1,646 m above sea level, is 1,959 kW. Three centrifugal water-cooled chillers (with a full-load efficiency of 0.63 kW/t and a capacity of 470 tons (1,653 kW) each) supply cold water for cooling the entire complex. Two small air coolers with piston pumps are used as backups for the centralized cooling system of the computer rooms.
Five central air conditioning units are equipped with 100% economizers for regulating the outside air temperature, heating heat exchangers, cooling heat exchangers and two parallel axial fans with adjustable blade pitch and a capacity of 56 kW each. Two units are equipped with systems for the recovery of heat from the air coming from the laboratories.
All major equipment in the complex, including mechanical systems, elements of the central air conditioning system, the complex’s internal and external lighting systems, etc., are connected to a building automation system (BAS). Direct digital control (DDC) technology is used to control all devices connected to the BAS. This system includes more than 4,600 data points. The operator uses a graphical interface to control the system via software on the main computer. In addition, the electrical distribution system is equipped with approximately 100 energy monitoring devices that detect faults in the electrical grid, track harmonic voltage disturbances, and measure and record electrical consumption. Each device monitors at least 20 parameters of the electrical system, including the power consumed at any given moment and the total energy consumption. These devices are networked to a central operator terminal to monitor the system’s status, issue warning signals, and record load curves. And these two systems form what the authors call a “building information system.”
During the design of the complex building, the engineering, architectural, and commissioning team provided GSA with significant initiative to reduce the complex’s construction and operating costs by implementing a number of highly effective measures to reduce the complex’s water and power costs. One of these measures included a new approach to commissioning the building based on the specifics of the construction work. Key activities included: pre-startup inspection of installed equipment, equipment start-up, testing, and verification of settings and operation, which were performed for all major mechanical and electrical equipment, fire and safety systems, BAS controls, and water-saving restrooms. In accordance with ASHRAE Guideline 151996: Commissioning Process for Heating, Ventilating, and Air-Conditioning Systems, the commissioning process included detailed training and system operation manuals. Following the completion of the construction phase, the focus shifted to commissioning after the start of operation of the complex and optimisation of the internal systems of the complex (POCx), which included:
— ensuring reliable and uninterrupted operation of all major systems of the complex;
— identifying any problems or malfunctions and finding solutions to eliminate them;
— finding and identifying opportunities to increase the efficiency of utility systems and overall improvement of equipment performance;
— development of the basis of documents regulating corrective actions during the operation of the system and actions aimed at improving its operation;
— creation of standards for managing the operation of the system in the future.
The general goals of the work carried out and POCx included: a) increasing the comfort of conditions in the complex (temperature and air quality in its premises); b) reducing energy consumption and increasing environmental safety; c) reducing the number and cost of maintenance work; d) increasing the service life of equipment.
The ability to monitor the operation of the power distribution network was expected to be useful for the planned work. Unfortunately, the monitoring system was initially only partially operational, allowing it to be used for only a few specific measurements. This system was later made fully operational, allowing the facility's staff to monitor the operation of the electrical systems.
Over the past decade, authors have used the term Post5Occupancy Commissioning (POCx) to identify activities performed after a facility has begun to be occupied, as opposed to activities performed during construction. While the authors acknowledge and support the view that proper commissioning should include both design and construction activities and post-occupancy activities, many commissioning projects often end after the project has been accepted and do not include post-occupancy activities.
These goals are consistent with those outlined in ASHRAE's Guideline 151996: Commissioning Process for Heating, Ventilating, and Air-Conditioning Systems and the more recent Portland Model Commissioning Plan and Recommended Specifications, Rev. 2.05.Energy Conservation, Inc., in cooperation with the Federal Energy Management Program. (When this material was written, ASHRAE's Guideline 052005: Commissioning Process was not yet complete.)
Methods
The primary strategy for POCx was to use existing utility information systems to monitor, record, and analyze the performance of the facility's mechanical systems, and then use that information to improve the facility's overall performance. The primary tools used to accomplish this were a BAS and a spreadsheet program. A centralized energy monitoring system would also have been useful, but it was not fully implemented until after POCx.
Of the more than 4,600 data points to which the BAS was connected, approximately 200 were identified as key data points — they determined the operating parameters of the plant's major mechanical systems. These key data points were continuously collected and trended every 15 minutes. The data was automatically stored on the central operator's terminal. Once a week, the data was downloaded over the network to the commissioning contractor for analysis. The major systems monitored included the boilers, chillers and associated ancillary equipment, five large air handling units of the central air conditioning system and a number of other ancillary systems.
The authors developed several semi-automated programs to generate spreadsheets from the trend analysis data imported from the BAS, plot graphs from combinations of these data, and create diagnostic tables summarizing the data. These graphs and tables were then analyzed manually to verify system performance and identify areas for improvement. They were used to continuously record system performance and efficiency. This allowed important events in system operation, such as subsystem and equipment failures, to be identified, which could then be used to create benchmarks for determining system performance.
Any equipment problems or malfunctions were recorded in a special table throughout the commissioning process of the complex. After the actions to eliminate them were taken, the results of these actions were monitored. On-site checks helped to confirm the elimination of the identified problems before starting to analyze the next points in the compiled tables.
 |
ОАТ — Outside temperature.
CH2 Amps — Current strength in the main cooler, A СНЗ Amps — Current strength in additional cooler, A CHWST — Chilled Water Temperature at Chiller Outlet. Lag Chiller Over-Cycling — Unnecessary frequent switching on and off of the auxiliary chiller. |
Figure 1. Chiller Schedule (June 9)
Results
The analysis revealed over twenty major problems and opportunities for improvement in the system, about 50% were related to the system operation and management, 25% were related to the system design, and another 25% were caused by equipment failures. This is a detailed example of just one problem related to the system operation. However, similar diagnostics and analysis were performed and documented for all other problems identified.
Too frequent switching of the auxiliary cooler operating modes.
As shown in Fig. 1, the averaged graph of the cooler operating parameters (created based on automatically collected data) shows that the additional cooler was frequently switched on and off — sometimes up to 4 times per hour and up to 20 times per day. Switching the operating modes of the cooler too frequently leads to its rapid wear, increases energy consumption, increases the cost of equipment maintenance and can quickly lead to the cooler failure. Based on the analysis of the obtained graphs, it was found that the additional cooler was switched off when the current in both coolers (CH2, CH3) dropped below
50% of the value corresponding to the operation of the coolers at maximum load. This value of the operating parameter setting corresponded to that specified in the design documentation created by the developer of the control system. Probably, the developer assumed that the base value would be adjusted during commissioning, but this was not done. Since the load in the system usually slightly exceeded the capacity of the main cooler, then after switching off the additional cooler, the water temperature (CHWST) increased, and the additional cooler was switched on again. The problem was solved simply: the threshold current value for disabling the additional cooler was reduced from 50 to 45% of the current at maximum load.
 |
OAT — Outside temperature.
CH2 Amps — Current in the main cooler, A СНЗ Amps — Current in the additional cooler, A CHWST — Chiller leaving water temperature. Lag Chiller Comes On When Secondary CHW Supply Temp. Rises — The secondary chiller comes on when the Lag Chiller Goes Off When One chiller Can Handle the Load — The secondary chiller goes off when the Lag Chiller Comes Back On, But With a Single On/Off Cycle — The lag chiller comes back on after |
Figure 2. Chiller Schedule (June 17-18)
Figure 2 shows the results of this change. If the load exceeds the capacity of the primary chiller, the water temperature rises and the secondary chiller is turned on. If the current in both chillers drops to 45% of the current at maximum load, the secondary chiller is turned off and remains off until the cooling capacity of the primary chiller is again exceeded. As Figure 2 shows, there was only one unnecessary shutdown of the secondary chiller on June 18, the cause of which cannot be determined without analyzing additional information. This means that the operation of the chillers should be monitored further (and perhaps with more frequent measurements) to identify any additional problems that may need to be addressed.
An inspection of the chiller control program code also revealed several timers and other variables associated with chiller control that were not directly controllable by the BAS operators. In the interest of managing the facility's utility systems, it was necessary to develop a simpler, more visual interface for monitoring and controlling chiller operation. The BAS company subsequently redesigned the chiller control graphical interface from the BAS central computer to display and control all of the variables needed to monitor the chiller system. This greatly improved the operators' ability to monitor the central refrigeration system.
Analysis methods similar to the above were developed for all of the other major mechanical systems.
Here are two more examples.
Frequent shutdowns of central air conditioning units due to static pressure build-up. The supply fans of the central air conditioning units frequently shut down due to low static pressure on the supply side of the axial fans. The fans remained off until the technical staff manually reset the system sensors. This could disrupt the required comfort level in the complex and cause an imbalance in the operation of the ventilation system. After identifying the cause of the fan shutdown, a temporary solution was found to the problem. The air valve control was switched to economizer mode (100% use of outside air) by directly changing the settings on the BAS central computer. A completely logical temporary solution for spring and autumn, it led to excess energy consumption and increased the load on the refrigeration system on warm days. But in the absence of another solution, this could also lead to the inability to provide the necessary air cooling in the summer heat.
Using the BAS graphs and observational data, the root cause of the problem was identified. It was an airflow control algorithm that, when the economizer mode was disabled (main outside air dampers closed, auxiliary dampers open), the recirculated air dampers were closed to maintain a minimum flow of outside air through the auxiliary damper. This was expected to close the supply fan inlet to increase the negative static pressure and
increasing the minimum outside air flow. However, the required negative static pressure was significantly lower than the safety sensor trigger point, which caused the fans to shut down (the minimum inlet size of the outside air supply valve was relatively small and probably too small for the system to operate properly).
The valve control algorithm was modified to include a minimum time interval for the main outside air supply valve. This eliminated unnecessary triggering of the safety sensors and also made it possible to meet the minimum ventilation level requirement while reducing fan power consumption.
Duct static pressure too high. When the facility was first in operation, the fan blade angles that pump air into the central air conditioning system were adjusted to maintain duct static pressure around 1.5 inches of water column (373.2 Pa). In our experience, this is the starting point for subsequent pressure adjustments.
Since variable air volume (VAV) units are nominally designed to operate in ducts with static pressures of 0.5 inH2O (124.4 Pa) or lower, an experiment was conducted to measure how much energy could be saved by reducing the duct static pressure and to see if occupant comfort would be compromised. At the time of the experiment, the electrical monitoring system had not yet been installed and portable current meters were installed in two central air conditioning units. During the experiment, fan power consumption, duct static pressure, outside temperature, and wind speed were recorded, first at a nominal 1.5 inH2O (373.2 Pa) duct static pressure and then at a nominal 1.0 inH2O (248.8 Pa).
The results showed that the expected savings from reducing fan power consumption was approximately 10% per year (i.e. $3,600), but there was no disruption to the VAV air handling unit (e.g., maximum airflow under standard terminal conditions) or to occupant comfort in the facility. As a result of this experiment, the baseline duct static pressure was reduced to 1 in. H2O (248.8 Pa) for all five central air handling units (and could be further reduced in the future).
Other results obtained during POCx included:
— detection and correction of excessive auxiliary boiler switching, inaccurate chilled water temperature control (overheating), faulty air temperature control system in central air handling units, incorrect air flow measurement system calibration values, poorly secured valve, and inaccurate or faulty BAS sensors and actuators;
– development of recommendations for improving the control algorithms of the system operation, including static pressure accumulation, economizer mode switching, heat recovery system operation, and cooling tower bypass valve operation.
After completion of POCx, several organizations conducted an energy use audit in accordance with US federal laws.
Prior to the testing, a centralized energy monitoring system was installed with sufficient resolution to measure actual grid load patterns, peak energy consumption, and the energy consumption of the various major end-user categories. Figure 3 shows the cumulative results of the three-day test. These data, together with others, can be used to develop benchmarks against which the system's performance can be continually compared under different weather conditions and different overall conditions, with the information systems properly configured and implemented.into the routine practice of managing the operation of the complex's utility systems.
Fig. 3 Shares of various categories of end consumers in daily consumption
of electricity (average data for August 24–26).
Computers, Lab Equipment, and Misc. Plug Loads – Computers, Lab Equipment, and Misc. Plug Loads,
plugged into electrical outlets. Cooling — Cooling. Lighting — Lighting.
HVAC Fans and Pumps — Fans and pumps for the
ventilation, heating, and air conditioning system.
Average outside temperatures for the three-day period under consideration:
Max: 90.5o F (32.5o C)
Min: 63.2o F (17.3o C)
Avg: 76.2o F (24.55o C)
Average daily energy consumption = 29,782 kWh/day
Average peak power consumption = 1,537 kW
Discussion
The work carried out at the POCx Research Center has improved the efficiency of its utility systems, simplified their maintenance, and increased the service life of the equipment. Some of the consequences are very difficult
to quantify. And given the above facts, a number of reasonable estimates of the economic efficiency of the work carried out can be made. In some cases, these estimates can be quite accurate. However, a significant portion of the benefits obtained cannot be directly quantified.
They can be described qualitatively or in the context of their potential impact on the operating costs of a utility system.
At the David Skaggs Research Center, annual utility costs, including electricity, natural gas, and water, were projected to be $19.16/m2. This figure was constructed using a computer model of the facility under construction, taking into account utility costs for the first three months after occupancy. The facility’s projected annual O&M budget is $12.16/m2. Funds for major repairs and replacements are included in the GSA budget as a separate line item. And the total cost that could be impacted by the POCx (excluding major repairs and replacements) is $31.32/m2 per year. If the expected cost savings are 5 to 10% (a conservative estimate given the literature, and further supported by the measured cost savings for the facility in question), the projected savings range from $1.40 to $2.90/m2.
An example of a benefit that is not quantifiable is the elimination of the excessive switching of the auxiliary boiler and auxiliary chiller described above. It is difficult to accurately quantify the excess wear and tear on the equipment, the additional maintenance costs associated with it, and the inconvenience to the people in the facility caused by the frequent switching of the equipment. It is conceivable that this problem would have been identified and corrected over time through normal operation and maintenance of the equipment, but the main benefit of POCx is that it was identified and corrected much earlier.
The cost of running POCx at the David Skaggs Research Center for six months was approximately $1.08/m2, or approximately 12% of the cost of all commissioning work performed during construction. This cost included only the cost of obtaining third-party consulting services prior to the scheduled work, the cost of weekly data analysis, on-site visits when needed, and interaction with the center's technical staff and contractors to troubleshoot any issues. Given the expected savings discussed above, this work will pay for itself in direct cost savings alone in less than one year.
The Future of Utility Information Systems
After six months of POCx, further work on this type of work was discontinued as there was no easy way to integrate it into the daily operations of the facility, requiring several hours of technical staff time per week. Technological advances needed to make more widespread use of such diagnostic techniques include:
– systems for automatic data collection and analysis that can integrate information from multiple independent control and measurement systems and then subject it to further analysis;
– the ability to create interactive graphs and reports that allow key information about the operation of utility systems to be quickly visualised, such as peak and daily energy consumption graphs,
etc.;
– an easy-to-use interface for running POCx, allowing technicians to easily create and use new diagnostic functions, graphs and reports as existing utility systems change and as the information requirements for managing those systems change.
These and other opportunities are gradually being offered by public, private organizations, government agencies, traditional automation system manufacturers, and new high-tech companies. The question is not whether new technologies will become more accessible and more widely used, but how quickly. As investments are made on an experimental basis, and the value of the necessary information compared to the costs of obtaining and analyzing it becomes more and more obvious, new technologies, which are more cost-effective, are one of the
main drivers of this market.
As the adoption and use of utility information systems increases, the costs of their creation and operation should decrease significantly. Over time, the use of such systems for the construction, commissioning, and commissioning of buildings will become the rule rather than the exception.
Even in properly constructed buildings, there is scope for continuous improvement in the performance of utility systems. Utility information systems promise to facilitate cost-effective maintenance, help improve comfort efficiency, and upgrade systems as they age and as equipment is replaced.
Authors: Mark Bowman, formerly a lead designer at E-Cube, is now a principal engineer at Bowman Consulting in Seattle. Jack S. Wolpert, president of E-Cube in Boulder, Colo., Ph.D., co-authored ASHRAE Guideline 1-1996: Commissioning Process for Heating, Ventilating, and Air-Conditioning Systems. Member of GPC Committee 14P, «Measuring Energy Demand and Consumption Reduction.»
Translated with permission from ASHRAE Journal, June 2006.
(c) American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. All rights reserved. Translated with abbreviations and distributed by BIG_RU Association. ASHRAE is not responsible for the accuracy of the translation.