Printer Friendly

Improvements to a methodology for estimating potential energy savings from existing building-commissioning/retrofit measures.


Today, as energy prices increase, saving money on energy bills through an existing building commissioning (EBCx) or an energy retrofit project is attractive to many commercial building owners. At the beginning of such a project, some form of screening is often applied to determine whether there is sufficient potential for savings to justify an EBCx assessment or an energy audit. If screening results are positive, the assessment/audit is performed and the potential for energy savings in the building is evaluated before the owner/operator decides that further work is likely to produce significant energy savings meeting the owner's economic criteria.

A popular technique that is used to screen for savings potential in a building is to compare its energy use per square foot of gross area to a group of buildings of similar type in the same climate. This technique is also known as energy benchmarking. Although this technique is very easy to use when a satisfactory database is available and gives some idea about the relative efficiency of the building, buildings are not always as similar as they appear. The buildings used for comparison are not necessarily energy efficient in general, and it gives no indication of energy conservation measures (ECMs) that merit consideration in the subsequent EBCx or retrofit process. Some of the improved energy benchmarking methods found in recent studies (Mills et al., 2008; Mathew and Mills, 2008; Yalcintas, 2006 and Cipriano et al., 2009) show potential in suggesting ECMs, but the other limitations still apply. Various energy simulation tools are also available to energy engineers. They can be used to predict savings from implementation of certain ECMs by changing inputs and comparing results. However, they are not designed to project the potential of savings in a building without detailed information about the building and the built-in system; in addition, they are usually complicated to use. Consequently, it would be desirable to have a methodology that is capable of predicting the opportunities for savings from low-cost/no-cost measures independent from the energy performance of other buildings. And, yet, this methodology should be easy enough to use in the early phase of EBCx assessments or energy audits to help decide if a comprehensive assessment should be carried out to identify commissioning measures or ECMs for further analysis.

Baltazar-Cervantes (2006) proposed such a methodology for estimating the potential energy savings in commercial buildings. At its core is a procedure for minimizing the energy cost required to maintain indoor thermal comfort. This methodology was applied to several existing buildings that have been retrofitted and/or commissioned. The measured savings in one of the buildings was about 85% of the estimated potential savings, close enough to suggest value for this approach. This methodology is promising, but to make it a useful tool in EBCx assessments or energy audits, further testing and improvement of the methodology is needed.


Baltazar-Cervantes's Methodology

Baltazar-Cervantes's (2006) methodology defines the potential energy savings in each outside air temperature bin as the difference between the actual energy cost during a particular period, preferably a whole year, and the minimum energy cost needed to maintain comfortable indoor conditions using the existing air-side HVAC systems in the building under the same weather conditions (Equation 1). Here the minimized energy cost is comprised of individual costs of electricity, cooling, and heating. The electricity cost consists of two parts: (1) lighting and equipment consumption, which is estimated from measured data and remains constant, and (2) fan power consumption, which is simulated (Equation 2).

Potential Energy Savings = Energy [Cost.sub.ACTUAL] - Energy [Cost.sub MINIMIZED] (1)

Energy Cost = (ELE [Cost.sub.LTEQ] + ELE [Cost.sub.FANP]) + CHW Cost + HHW Cost (2)

The binned outside air dry-bulb temperature data, mean coincident humidity ratios for each bin, and the measured energy consumption data needed to determine the potential savings can be prepared from hourly measured weather and consumption data.

The essence of this methodology is the procedure for determining the minimum energy use cost, which has two major components, as Figure 1 demonstrates: the model is illustrated as a compound function in the figure, which thermodynamically represents the performance of the built-in HVAC system and the numerical procedure for energy cost minimization. The model calculates weather-dependent loads that are input to an air-side system simulation. Both the load calculation and system simulation follows the modified bin method (Knebel 1983). The numerical procedure generates and seeks the parameter values that will produce minimum total energy cost while meeting the indoor thermal comfort requirements.


Sequential exhaustive search is employed as the optimization method within each set of binned ambient conditions. Figure 2 illustrates the procedure for implementing the methodology to determine the minimum energy cost for each bin. The total potential energy cost savings during the period evaluated are then the sum of savings found for all bins.


Improvements and Discussion of Methodology

The improved methodology is implemented in a spreadsheet-based prototype program called the Potential Energy Savings Estimation (PESE) Toolkit to test the methodology. This section presents the major improvements implemented and discusses the methodology.

Optimization Parameters. In Baltazar-Cervante's (2006) implementation of the methodology, four parameters are selected for optimization: cold deck and hot deck (for dual-duct systems) leaving air temperature setpoints, minimum supply airflow per square foot of floor area (for variable-air-volume [VAV] systems), and the outside airflow as a fraction of total design airflow. In this study, the volumetric outside airflow is optimized instead of optimizing outside air fraction because volumetric control is required to implement the optimization result; the minimum supply airflow is not optimized since the optimized value is always equal to the designated lower limit. In addition to the above changes, room temperature setpoints in the exterior and interior zones are included as additional optimization parameters, since space loads are dependent on these two parameters.

In summary, five parameters are selected for optimization in this study: exterior and interior zone room temperature setpoints, cold deck and hot deck leaving air temperature setpoints, and outside airflow rate. In addition, in the methodology implemented in the PESE toolkit, options are provided to users to optimize any combination of these five parameters. This is helpful in evaluating savings based on the existing control capability. For example, Baltazar (2006) noted that his methodology assumed an economizer and predictably seemed to over-estimate savings at lower outside air temperatures in buildings that do not have an economizer. In such cases, the user can choose to estimate potential savings based on the current system setting, or to determine the extra savings achievable by installing an economizer.

Limits on Optimization Parameter Values. To make the optimization result useful, it is important to set appropriate lower and upper limits on the values of optimization parameters. These limits should be determined based on the special requirements of each application. However, the considerations used to determine the limits in the case study are given here for reference.

Room temperature setpoints. ASHRAE's general design criteria for commercial and public buildings can be adopted when there is no specific requirement for room temperature and relative humidity control. For example, 70[degrees]F-78[degrees]F (21.1[degrees]C-25.6[degrees]C) and up to 60% RH are acceptable for offices (ASHRAE 2007a) during occupied periods; 65[degrees]F-85[degrees]F (18.3[degrees]C-29.4[degrees]C) and up to 70% RH can be used as reset values during unoccupied periods.

Cold deck and hot deck leaving air temperature setpoints: Limits on these setpoints usually vary from project to project. Following common EBCx practice in hot and humid climates, the reset ranges can be 55[degrees]F-70[degrees]F (12.8[degrees]C-21.1[degrees]C) for cold deck and 70[degrees]F-110[degrees]F (21.1[degrees]C-43[degrees]C) for hot deck temperatures.

Outside airflow rate. Minimum outside air supply in the breathing zone required by ANSI/ASHRAE Standard 62.1-2007 (ASHRAE 2007b) can be adopted as a lower limit. For example, 5 cfm/person (2.4 L/s * person) and 0.06 cfm/[ft.sup.2] (0.31 L/s * [m.sup.2]) are generally required in offices; however, the outside air requirement can be only 7 cfm/person (3.3 L/s * person) with no minimum requirement on cfm/[ft.sup.2] when a [CO.sub.2] sensor is available to maintain a [CO.sub.2] level of 1000ppm.

Minimum Airflow Setting. Exterior and interior zone minimum airflows are not parameters to be optimized by the methodology employed in this study. However, resetting minimum airflow is an important commissioning measure in VAV systems and usually has significant influence on the energy use.

In EBCx practice, the minimum airflow should be checked and reset if necessary for each individual VAV terminal box. This requires knowledge of the loads in each space as well as design information and terminal box details. Since this methodology is developed to assist in the early stages of an EBCx or energy audit process, the above information is usually not available and not much effort can be expended to determine minimum airflow. According to Taylor and Stein (2004), ANSI/ASHRAE Standard 62.1-2007 for ventilation, and ANSI/ASHRAE Standard 90.1-2007 for energy (ASHRAE 2007c), the minimum airflow during the occupied period can be reset to the largest of the following: (1) the airflow required to meet the design heating load at a supply air temperature that is not too warm (e.g., 85[degrees]F [29.4[degrees]C]), (2) 30% of design airflow or 0.3 cfm/[ft.sup.2] (1.5 L/s * [m.sup.2]) if the design airflow is oversized, or (3) the minimum breathing zone outside air required by ANSI/ASHRAE Standard 62.1-2007. The minimum airflow during the unoccupied period can be reset to zero in many cases.

Space Load Calculation. In Baltazar-Cervantes's (2006) implementation, space cooling and heating loads are calculated based on fixed occupied period room temperature setpoints (e.g., 75[degrees]F [23.9[degrees]C]). This can lead to inaccurate optimization results when the room temperatures are optimized using unoccupied resets and seasonal resets because the conduction load makes up a significant fraction of the total space load and is proportional to the difference between the room temperature and the outside air temperature. In office buildings, for example, room temperature can have a relatively wide acceptable range: 70[degrees]F-78[degrees]F (21.1[degrees]C-25.6[degrees]C) during occupied periods and 65[degrees]F-85[degrees]F (18.3[degrees]C-29.4[degrees]C) during unoccupied periods. Therefore, in this study, a space load calculation procedure is developed based on the modified bin method and linked with the optimization procedure, so the space load will be recalculated dynamically as room temperature setpoints change in the optimization process.

Simulation of Buildings with Multiple Types of Systems. Many buildings have more than one type of system. To make the methodology applicable to such buildings, the fractions of exterior and interior zone areas served by each type of system are applied to calculated whole-building exterior and interior zone space loads. It is assumed that the space load is proportional to floor area. This assumption works well for buildings having each type of system serving an entire floor or several floors, or buildings having two different types of systems serving the exterior zone and interior zone, respectively.

Air-Handling Unit (AHU) Shut-Down Simulation. Shutting down the AHU(s) during unoccupied periods is a common and effective ECM. When the AHU is turned on before the building is occupied again, it has to bring the room temperature to setpoint in a short time. Observations of the measured consumption data show that the cooling or heating energy consumption during the start-up period is usually significant. In addition, during a shut-down period, the AHU will normally be turned on if a lower or upper limit on the room temperature is reached. Therefore, the cooling and heating energy use during the unoccupied period needs to be estimated in a reasonable manner. This energy use can be estimated to be approximately equal to the sum of the largest two components of the space load: the internal heat gain and the conduction load.

During the AHU shut-down period, the room temperature changes under the influence of internal heat gain and conduction through the building envelope. As a result, the conduction load can be significantly different from that when the room temperature is kept at the occupied period setpoint. This challenges one of the major limitations of the modified bin method, which is based on time-averaging techniques and does not take the thermal capacitance of the space into account. However, based on the measured data in an office building in Texas, where AHU shutdown has been implemented, it is found that the average room temperature during the unoccupied period has an approximately linear relationship with the average outside air temperature. This finding is used to estimate the average conduction load during the unoccupied period. It is noted that the relationship can vary from building to building depending on the building's size, construction, internal heat gain, etc. Nevertheless, the relationship obtained is used as the default in the methodology implemented in the PESE Toolkit; this relationship can be modified based on engineering considerations as warranted.

Air-Side Simulation Models. Air-side simulation models for four common HVAC systems are included: single-duct and dual-duct constant volume systems, as well as single-duct and dual duct VAV systems. The models employed in Baltazar-Cervantes's (2006) methodology largely came from the modified bin method, which was developed by ASHRAE TC4.7 and described in Knebel (1983). These models are used in this study with several modifications to better represent the performance of the systems for the conditions simulated.



The improved methodology implemented in the PESE Toolkit was tested in a single-story building with a total area of 19,363 [ft.sup.2] (1799 [m.sup.2]). The building consists of a large display hall, offices, a small library, and a conference room. It is generally open between 8:00 a.m. and 5:00 p.m., Monday through Friday, except on holidays. The HVAC system is a single-duct variable-air-volume (SDVAV) system with terminal reheat VAV boxes. One air-handling unit (AHU) serves the entire building. Implementation of EBCx measures was completed in this building on November 2, 2007.

The current energy use is first simulated and the simulation calibrated to measured energy use. Next, energy savings are estimated using single-parameter optimizations, i.e., optimizing one applicable parameter at a time, for better understanding of the optimized profile of parameters and energy uses. Finally, potential energy savings are estimated by a performing multiple-parameter optimization, i.e., optimizing all the applicable parameters together.

Simulation without Optimization

The basic building information including dimensions, internal heat gain, envelope characteristics, etc., required in the simulation with the PESE Toolkit is collected by on-site investigation. Information about the HVAC system is obtained from the energy management system (EMS) with some parameters remaining uncertain. The subsequent single-parameter optimization requires that the values of these parameters be determined, which is accomplished using the method of calibrated signatures developed by Wei et al. (1998). The "calibrated" input values used in the simulation are given in Figure 3, which shows the input interface of the PESE Toolkit. One year of hourly weather and measured consumption data for December 1, 2007, through November 30, 2008, (post-EBCx) are sorted into 5[degrees]F (2.8[degrees]C) bins with occupied and unoccupied periods distinguished, as shown in Figure 4. No parameter is selected for optimization at this point. The simulated annual energy use and costs after calibration are given in Table 1 as the baseline for optimization.


Name Sanders Corps of Cadets Center
City College Station
Latitude 30[degrees]
Orientation (Wall# 1) NE
Length 180 ft
Width 110 ft
Height 15ft
Above ground floors 1
Has Basement FALSE
Basement conditioned FALSE
[A.sub.tot] 19,800 [ft.sup.2]
[A.sub.e] 7,800 [ft.sup.2]
[A.sub.i] 12,000 [ft.sup.2]

 Internal Heat Gain

[Ocp.sub.e] 20 pep
[Ocp.sub.i] 20 pep
AveOcpFactor (Ocp) 1.00
AveOcpFactor (Unocp) 0.10
LTEQ (Ocp) 54 kW
LTEQ (Unocp) 25 kW


FPS_July 0.72
FPS_January 0.48
Tpc 107 [degrees]F
Tph 27 [degrees]F
[T.sub.o,des] 86 [degrees]F


U-wall 0.09 Btu/(h * [ft.sup.2] * [degrees]F)
U-window 1.00 Btu/(h * [ft.sup.2] * [degrees]F)
U-roof 0.05 Btu/(h * [ft.sup.2] * [degrees]F)
U-g round 0.05 Btu/(h * [ft.sup.2] * [degrees]F)
[]/[A.sub.wall] 1 25.0%
[]/[A.sub.wall] 2 15.0%
[]/[A.sub.wall] 3 25.0%
[]/[A.sub.wall] 4 15.0%
[A.sub.skylights] 0 [ft.sup.2]
SC 1 0.45
SC 2 0.15
SC 3 0.20
SC 4 0.25
SC skylights 0.00
Wall color Medium colored
Roof color Dark colored


System Type SDVAV

Reheat Type Hot Water

Preheat Type Hot Water

Has Economizer FALSE

AHU shut off FALSE

Fraction of [A.sub.e] 100.0%

Fraction of [A.sub.i] 100.0%

Zone T set point during occupied hours

[T.sub.e_ocp] 72

[T.sub.i_ocp] 72

Zone T reset point during unoccupied hours

[T.sub.e_unocp] 72

[T.sub.i_unocp] 72

[V.sub.TD] 27,545 cfm

[V.sub.e] cfm

[V.sub.i] cfm

[T.sub.CL] (setpoint 64 @ [T.sub.OA]1= 25
1) [degrees]F [degrees]F

[T.sub.CL] (setpoint 57 @ [T.sub.OA]2= 65
2) [degrees]F [degrees]F

[T.sub.HL] (setpoint [degrees]F @ [T.sub.OA]1= [degrees]F

[T.sub.HL] (setpoint [degrees]F @ [T.sub.OA]2= [degrees]F

[P.sub.SF-rated] 30 hp [[eta].sub.SF] 1

[P.sub.RF-rated] 0 hp [[eta.sub.RF] 1

OA controlled by [VOA.sub.max] 4,500 cfm

[X.sup.OA,min_ocp] [X.sup.OA,min_unocp]

[V.sup.OA,min_ocp] 700 cfm [V.sub.OA,min_unocp] 500 cfm

 VAV systems

VAV mechanism Variable Speed Drive

[V.sub.e,min_ocp] 7,200 cfm [V.sub.e,min_unocp] 6,370 cfm

[V.sub.i,min_ocp] 11,080 cfm [V.sub.i,min_unocp] 9,800 cfm


Variables Select Ocp:range&grid Unocp:range&grid

Te FALSE ([degrees]F) 70 - 78 9 65 - 85 11
Ti FALSE ([degrees]F) 68 - 72 5 65 - 85 11
TCL FALSE ([degrees]F) 55 - 70 16 55 - 70 16
THL FALSE ([degrees]F) 80 - 115 16 80 - 115 16
VOA FALSE (cfm) 600 - 4,500 11 0 - 4,500 12

ELE Price 0.092 $/kWh
CHW Price 9.602 $/MMBtu
HHW Price 13.099 $/MMBtu
RHz1 10%
RHz2 60%

Figure 3 Building information, system information, and optimization
options input in the calibrated simulation.



Outside air Outside air Zone relative Hours of
temperature humidity ratio - humidity (%) occurrence (hr)


 27 0.001948 65.2 2

 32 0.002618 69.5 5

 37 0.003551 77.1 33

 42 0.004116 73.4 56

 47 0.004348 64.1 110

 52 0.005150 62.9 112

 57 0.005589 56.8 163

 62 0.006464 54.9 192

 67 0.008619 61.3 290

 72 0.010431 62.3 351

 77 0.012683 63.9 382

 82 0.014488 61.8 374

 87 0.014880 54.0 290

 92 0.014615 45.4 282

 97 0.013313 35.5 140

102 0.012134 27.9 11

107 0.012752 25.2 1


 27 0.002106 70.5 11

 32 0.002643 70.1 76

 37 0.003461 75.2 186

 42 0.003995 71.3 303

 47 0.004601 67.8 378

 52 0.005675 69.3 428

 57 0.007088 71.9 445

 62 0.008834 74.8 538

 67 0.010686 75.7 646

 72 0.013023 77.5 813

 77 0.015470 77.6 1,030

 82 0.015346 65.3 532

 87 0.014747 53.5 298

 92 0.014614 45.3 205

 97 0.012826 34.2 82

102 0.010645 24.5 10



Measured ELE Measured CHW Measured HHW consumption
consumption (kWh) consumption (kBtu) (kBtu)

ELE_Meas CHW_Meas HHW_Meas

 89 241 410

 255 912 1,246

 1,998 6,122 7,843

 3,651 11,484 12,516

 7,204 25,093 21,783

 7,262 27,096 21,622

 10,795 41,775 29,424

 12,654 52,277 31,724

 19,161 85,079 42,959

 23,177 109,389 47,778

 25,705 130,655 44,621

 25,465 137,366 37,379

 20,511 116,138 25,384

 20,020 123,951 22,089

 10,176 66,872 11,207

 816 5,560 1,003

 75 494 106


 379 1,559 2,471

 2,742 13,073 18,875

 6,660 33,911 43,205

 10,867 59,504 63,529

 13,139 77,720 73,090

 14,643 93,132 78,959

 14,865 102,549 75,314

 18,285 132,524 81,251

 21,702 168,176 90,246

 27,466 231,191 95,715

 35,245 339,699 96,825

 17,628 190,299 47,773

 8,741 111,242 24,767

 5,884 78,678 16,114

 2,444 33,005 6,635

 302 4,001 846

Figure 4 Bin data input in the "calibrated" simulation.

Table 1. Baseline Annual Energy Use and Costs

 (kWh) (kWh)

Annual Occupied 186,789 1018 204 (59,787)
energy use (298,346)

 Unoccupied 204,418 1696 969
 (497,049) (283,986)

 Total 391,208 2714 1172
 (795,395) (343,479)


Annual Occupied $17,185 $9775 $2671 $29,630

 Unoccupied $18,806 $16,289 $12,686 $47,782

 Total $35,991 $26,064 $15,357 $77,412

Reset Minimum Airflow

Following the procedure described earlier, it is determined that the exterior and interior zone minimum airflow during the occupied period in this building is reset from 7200 cfm (3398 L/s) and 11,080 cfm (5229 L/s) to 3820 cfm (1803 L/s) and 3600 cfm (1699 L/s), respectively, and the minimum airflow during the unoccupied period is reset from 6370 cfm (3006 L/s) and 9800 cfm (4625 L/s) to zero. These reset values are used in the following single and multiple parameter optimizations. Figure 5 compares the cooling, heating, and fan power energy use before and after resetting the minimum airflow. The result shows that the savings that can be achieved from this minimum flow reset alone are very significant: total reductions of 20% and 54% during occupied and unoccupied periods, respectively, as shown in Table 2.
Table 2. Annual Savings from Resetting the Minimum Airflow,
Single-Parameter Optimization, and Multiple-Parameter Optimization

 ELE CHW Savings

Occupied Only reset minimum flow $1166 7% $219 22%

 [T.sub.e], [T.sub.i] $1782 10% $2909 30%

 Single-parameter optimization $1494 9% $2334 24%

 [V.sub.OA] $1166 7% $2795 29%

 Multiple-parameter $1964 11% $3525 36%

Unoccupied Only reset minimum flow $3990 21% $9962 61%

 [T.sub.e], [T.sub.i] $4932 26% $13,502 83%

 Single-parameter optimization $4169 22% $10,030 62%

 [V.sub.OA] $3990 21% $10,943 67%

 Multiple-parameter $4875 26% $14,381 88%

 HHW Savings Total Savings

Occupied Only reset minimum flow $2432 91% $5796 20%

 [T.sub.e], [T.sub.i] $2536 95% $7228 24%

 Single-parameter optimization $2516 94% $6344 21%

 [V.sub.OA] $2431 91% $6392 22%

 Multiple-parameter $2565 96% $8054 27%

Unoccupied Only reset minimum flow $11,651 92% $25,603 54%

 [T.sub.e], [T.sub.i] $12,235 96% $30,669 64%

 Single-parameter optimization $11,729 92% $25,928 54%

 [V.sub.OA] $11,634 92% $26,567 56%

 Multiple-parameter $12,306 97% $31,562 66%



Single-Parameter Optimization

Four out of the five available optimization parameters in PESE are applicable to this building. To better understand the effect on the energy use from each category of these parameters independently, they are grouped as follows to be activated for optimizations: exterior and interior zone temperature setpoints ([T.sub.e] and [T.sub.i]), cold deck leaving air temperature ([T.sub.CL]), and outside airflow ([V.sub.OA]). The lower- and upper-limiting values of each parameter, as well as the number of grid divisions used in the optimization, are given in Table 3. The savings shown in Table 2 for each single-parameter optimization and multiple-parameter optimization represent the combined impact of optimizing the parameter(s) shown in addition to reducing minimum airflow; the same quantities are compared graphically in Figure 6. The profiles of optimized parameter settings as functions of ambient temperature during occupied and unoccupied periods are given in Figure 7. The optimized cooling, heating, and fan power energy use is compared with only resetting the minimum airflow in Figure 8.
Table 3. Optimization Parameter Limits

 Optimization Lower Limit Upper Limit Grid Divisions

Occupied [T.sub.e], 70 (21.1) 78 (25.6) 9

 [T.sub.i], 68 (20.0) 72 (22.2) 5

 [T.sub.CL], 55 (12.8) 70 (21.1) 16

 [V.sub.OA], cfm 600 (283) 4500 (2124) 11

Unoccupied [T.sub.e], 65 (18.3) 85 (29.4) 11

 [T.sub.i], 65 (18.3) 85 (29.4) 11

 [T.sub.CL], 55 (12.8) 70 (21.1) 16

 [V.sub.OA], cfm 0 (0) 4500 (2124) 12


The results reveal that single-parameter optimization during the occupied period can save only 1%-4% more than resetting the minimum airflows. This also holds true for the unoccupied period except that 10% additional savings are possible by optimizing the zone temperatures. This is expected since this building has already undergone the EBCx process, and these optimization parameters are among the typical EBCx measures.

Zone Temperature Set-Point Optimization

There are two main ways in which the exterior and interior zone temperature setpoint could affect the total energy cost: (1) minimizing indoor and outdoor temperature difference can significantly reduce space heating and sensible cooling loads. Figure 9 compares sensible space heating and cooling load before and after optimizing zone temperature setpoints; (2) when supply airflow is higher than the minimum value in cooling mode, a higher return air temperature requires smaller supply airflow and saves fan power on one hand but it increases cooling coil sensible load. The former dominates in the range selected for optimization since the electricity price is nearly three times the chilled-water price for the same amount of energy.

The optimized exterior zone temperature setpoint stays at the lower limit (70[degrees]F [21.1[degrees]C]), which reduces space heating load, when outside air temperature is lower than 57[degrees]F (13.9[degrees]C) and 52[degrees]F (11.1[degrees]C) during occupied and unoccupied periods, respectively. It gradually rises to the upper limit (78[degrees]F [25.6[degrees]C]) at the 82[degrees]F (27.8[degrees]C) bin and the 72[degrees]F (22.2[degrees]C) bin, respectively. The increased exterior zone temperature at higher ambient temperatures has reduced space-cooling load significantly.

The optimized interior zone temperature setpoint stays at its upper limit (72[degrees]F [22.2[degrees]C]) constantly during the occupied period; it increases from 67[degrees]F (19.4[degrees]C) at the 32[degrees]F (0[degrees]C) bin to its upper limit (85[degrees]F [29.4[degrees]C]) at the 57[degrees]F (13.9[degrees]C) bin during the unoccupied period, and the space cooling load is significantly reduced. During both periods, the interior zone supply airflow is always higher than the minimum value. Consequently, a higher temperature setpoint is preferred to save fan power, except for a few low temperature bins during the unoccupied period where the cooling load is light. The effect of reducing cooling coil sensible load from a lower return air temperature more than offsets the slightly increased fan power.

Cold Deck Leaving Air Temperature Optimization

The optimized cold deck leaving air temperature ([T.sub.CL]) decreases with increasing bin temperature in general during both occupied and unoccupied periods. There are three major principles driving this optimization result in both the exterior and interior zones: (1) in cooling mode at minimum airflow, reheating and sensible cooling energy will be saved if [T.sub.CL] is set higher; (2) in cooling mode when airflow exceeds the minimum, fan power will be saved if [T.sub.CL] is set lower; (3) in heating mode, reheat energy will be saved if [T.sub.CL] is set higher.

The optimized [T.sub.CL] reset profile during the occupied period gradually decreases from 64[degrees]F (17.8[degrees]C) to the lower limit (55[degrees]F [12.8[degrees]C]) at 67[degrees]F (19.4[degrees]C). At 62[degrees]F (16.7[degrees]C) and lower, the exterior zone airflow stays at its minimum which favors a higher [T.sub.CL], while the interior zone airflow exceeds minimum which favors a lower [T.sub.CL]. The optimized profile is the result of a balance between these two factors, and as the outside temperature goes lower, the second factor becomes more dominant. At 67[degrees]F (19.4[degrees]C) and higher, both exterior and interior zone airflows exceed their minimums, which keeps the optimized [T.sub.CL] at its lower limit (55[degrees]F [12.8[degrees]C]).

The optimized [T.sub.CL] reset profile during the unoccupied period also generally decreases from lower to higher temperature bins. At 27[degrees]F (-2.8[degrees]C), both the exterior and interior zones are in heating mode; thus, the optimized [T.sub.CL] goes to its upper limit (70[degrees]F [21.1[degrees]C]). As the ambient temperature increases to 57[degrees]F (13.9[degrees]C), the exterior zone is still in heating while the interior zone is in cooling and its airflow exceeds the minimum flow. The optimized profile is the result of a balance between the two factors. At 62[degrees]F (16.7[degrees]C) and higher, both zones are in the cooling mode, which brings [T.sub.CL] to its lower limit (55[degrees]F [12.8[degrees]C]).

Outside Airflow Optimization

The optimization result shows that as the ambient temperature increases, the outside airflow drops to its lower limit at 72[degrees]F (22.2[degrees]C) and 67[degrees]F (19.4[degrees]C) during occupied and unoccupied periods, respectively, where latent loads appear. It drops one bin earlier during the unoccupied period since the mean coincident outside air relative humidity is higher than during the occupied period. At lower temperature bins, making use of outside air for free cooling can significantly reduce mixed air temperature and save cooling coil sensible load. However, as outside air temperature gets lower, preheating energy will be required at maximum outside airflow. As a result, during the occupied period, the optimized outside airflow reaches its upper limit (4500 cfm [2124 L/s]) from 47[degrees]F-67[degrees]F (8.3[degrees]C-19.4[degrees]C), and decreases at lower temperatures to 2160 cfm (1019 L/s) (at bin 27[degrees]F [-2.8[degrees]C]); during the unoccupied period, it gradually increases from 400 cfm (189 L/s) (at 27[degrees]F [-2.8[degrees]C]) to 3300 cfm (1557 L/s) (at 62[degrees]F [16.7[degrees]C]) without ever reaching the upper limit because the total airflow is much smaller than during the occupied period.

Multiple-Parameter Optimization

A full-fledged optimization with all of the above four parameters activated is performed with the same parameter limits given in Table 3. The potential savings obtained from this multiple-parameter optimization are listed in Table 2 and illustrated in Figure 6 where they are compared with resetting minimum airflow and the single-parameter optimizations (combined with reset of minimum airflow). An extra 7% and 12% total savings during occupied and unoccupied periods, respectively, are possible in addition to the savings achieved from resetting minimum airflow. For all energy use categories, the savings from multiple-parameter optimization (ELE: 11% [occupied] and 26% [unoccupied], CHW: 36% [occupied] and 88% [unoccupied], HHW: 96% [occupied] and 97% [unoccupied]) are just slightly higher than the largest savings from single-parameter optimizations, which resulted from optimizing the zone temperatures. Figure 10 gives the profiles of optimized parameter settings during both the occupied and unoccupied periods. Figure 11 shows the optimized cooling, heating, and fan power energy use compared with only resetting minimum airflow.


In this study, Baltazar-Cervantes's (2006) methodology for potential energy savings estimation is improved in several ways and implemented in a spreadsheet-based prototype program. The improved methodology is expected to help pre-screening for potential savings in the early phase of existing building commissioning assessments or energy audits. The implemented methodology is tested in a building, and the results show that there is limited room for savings in this building beyond potential savings from resetting the minimum airflow. This is expected since the EBCx process had been applied to this building. The optimized profiles of the control parameter settings determined from the single-parameter optimizations can be explained with basic engineering principles, which supports the findings of the methodology.


ASHRAE. 2007a. 2007 ASHRAE Handbook--HVAC Applications. Atlanta, GA: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

ASHRAE. 2007b. ANSI/ASHRAE Standard 62.1-2007, Ventilation for Acceptable Indoor Air Quality. Atlanta, GA: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

ASHRAE. 2007c. ANSI/ASHRAE Standard 90.1-2007, Energy Standard for Buildings Except Low-Rise Residential Buildings. Atlanta, GA: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

Baltazar-Cervantes, J.C. 2006. Development of an automated methodology for calibration of simplified air-side HVAC system models and estimation of potential savings from retrofit/commissioning measures. Mechanical Engineering Department, Texas A&M University, College Station, TX.

Cipriano, X., J. Carbonell, and J. Cipriano. 2009. Monitoring and modeling energy efficiency of municipal public buildings: Case study in Catalonia region. International Journal of Sustainable Energy 28(1):3-18.

Knebel, D.E. 1983. Simplified Energy Analysis Using the Modified Bin Method. Atlanta, GA: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

Mathew, P., and E. Mills. 2008. Action-oriented benchmarking: Using the CEUS database to benchmark commercial buildings in California. Energy Engineering 105(5):6-18.

Mills, E., P. Mathew, and M.A. Piette. 2008. Action-oriented benchmarking: Concepts and tools. Energy Engineering 105(4):21-40.

Taylor, S.T., and J. Stein. 2004. Sizing VAV boxes. ASHRAE Journal 46(3):30-35.

Wei, G., M. Liu, and D.E. Claridge. 1998. Signatures of heating and cooling energy consumption for typical AHUs. Proceedings of the Eleventh Symposium on Improving Building Systems in Hot and Humid Climates, June 1-2, Fort Worth, TX, pp. 387-402.

Yalcintas, M. 2006. An energy benchmarking model based on artificial neural network method with a case example for tropical climates. International Journal of Energy Research 30:1158-74.

Jingjing Liu

Student Member ASHRAE

Juan-Carlos Baltazar, PhD


David E. Claridge, PhD, PE


Jingjing Liu is a graduate student in the Department of Mechanical Engineering, Texas A&M University, as well as a research assistant at Energy Systems Laboratory, College Station, TX. Juan-Carlos Baltazar is an associate research engineer at Energy Systems Laboratory, College Station, TX. David E. Claridge is Leland Jordan Professor in the Department of Mechanical Engineering, Texas A&M University, as well as director at Energy Systems Laboratory, College Station, TX.
COPYRIGHT 2011 American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Liu, Jingjing; Baltazar, Juan-Carlos; Claridge, David E.
Publication:ASHRAE Transactions
Article Type:Report
Geographic Code:1USA
Date:Jan 1, 2011
Previous Article:Determining the average R-value of tapered insulation.
Next Article:Methods for estimating heating and cooling degree-days to any base temperature.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |