1、4843 (RP-1197) Effect of Data Availability on Modeling of Residential Air Conditioners and Heat Pumps for Energy Calculations Michael R. Wassmer Associate Member ASHRAE ABSTRACT A component-based model for residential air conditioner and heat pump energy calculations has been developed through ASHRA
2、E Research Project 1197. A primary objective of this project was to develop a suitable model that requires only readily available, or potentially readily available, input data. The component model meets this objective by only requir- ing rating data that are generated from ARIStandard 21 01240. Unfo
3、rtunately, only a small percentage of the data collected during these tests is published by ARI. Some manufacturers releuse u more signijcuntportion of the test dutu in their tech- nical literature, but these data are often incomplete. This paper describes the development of default data for the mod
4、eling of residential air conditioners and heat pumps using a test suite of 25 units. The paper evaluates the impact of limited data availability, and the need to use default data, on modelpredic- tion accuracy. The results show that the use of minimal data open produces acceptable results, though ac
5、curacy improves sign$cantly with additional rating data. Of most importance are the power at standard rating conditions, the type of compressol; and the coil airflow rates. The results also show that accurate results are obtained without knowing the exact compressor model number: INTRODUCTION Comput
6、erized energy calculations are now routinely used for evaluating equipment alternatives during engineering design and specification. However, this application is less accurate and efficient than possible due to limitations in avail- able equipment simulation models and a mismatch between input requi
7、red by those models and the data published by Michael J. Brandemuehl, PhD, PE Member ASHRAE manufacturers. These shortcomings are particularly signifi- cant in residential energy analysis when it is desired to compare specific equipment alternatives in a particular build- ing and climate. A new mode
8、l for residential air conditioner and heat pump energy calculations was recently developed under ASHRAE Research Project 1197, Updated Energy Calcula- tion Models for Residential HVAC Equipment (Brandemuehl and Wassmer 2005). The new model, referred to here as the “component model,” is fundamentally
9、 different from the models currently residing in the most popular state-of-the-art whole building energy analysis programs. Almost ali current models use minimal equipment performance at standard conditions and regression equations based on the performance characteristics of typical equipment (e.g.,
10、 Winkelmann et al. 19931). The component model also requires equipment performance at standard conditions, but it uses a combination of first-principle models and empirical models of the primary system components to calculate the performance of the actual equipment at nonstandard operating condition
11、s. Its combina- tion of accuracy and minimal input requirements make the component model an updated alternative to the current models in building energy analysis programs. Application of any HVAC equipment model, including the component model, requires data on equipment character- istics. In the abs
12、ence of such equipment characteristics, users are forced to use default values to fill the voids. This paper explores the availability and quality of data required by the component model, the development of default data, the effects of poor rating data on the results, and suggestions for future rati
13、ng data standards to maximize the performance of energy simulation models and their ease of use. Michael R. Wassmer is an engineer at the National Renewable Energy Laboratory, Golden, Colo. Michael J. Brandemuehl is an associate professor in the Department of Civil, Environmental, and Architectural
14、Engineering, University of Colorado at Boulder. 21 4 02006 ASHRAE. COMPONENT MODEL The component model consists of a combination of first- principle and empirical models of individual direct expansion (DX) system components (Wassmer and Brandemuehl 2005). Given models of the compressor, condenser, e
15、vaporator, and expansion device, energy and mass balances are performed on the vapor compression cycle using simplified refrigerant prop- erties (Bourdouxhe et al. 1996). The compressor model is an empirically and statistically based correlation for positive displacement compressors. The model has p
16、olynomial fits for the refrigerant mass flow rate, mrer, and the power draw of the compressor, Pcomp. There are two independent variables in each ten-tem equation. They are the saturated suction temperature, SSZ and the saturated discharge temperature, SDT = M, +M2ssT+M3sDT+Mqss+M5(ssT)(sDT)+MgsD mr
17、ef; rated + M,SS? + M(SDT)(SS?) + M(SST)(SD?) + MSD? (1) camp = pl +P2SST+P3SDT+PqSS12+P5(SST)(SDT)+PgSD? + P,SS? + P,(SDTXSS?) + P,(SST)(SD?) + P,SD? comp,rafed (2) The rating conditions and form of the equations are spec- ified by AR1 Standard 540 (AR1 1999). Specifically, the rated flow rate and
18、power are determined at SST= 45F (7.2“C) and SDT= 130F (54.4“C). The coefficients are typically available directly from the manufacturers. The condenser is simulated using an air-to-liquid coil model that accounts for heat transfer between air and a liquid through a dry-finned coil surface. The mode
19、l is based on simple effectiveness-NTU heat exchanger relationships, described in most introductory heat transfer textbooks, and assumes that the temperature of the refrigerant throughout the coil is fixed at the SDT. The evaporator is also modeled using effectiveness-NTU relationships. For heat pum
20、ps, the same dry coil model is used. For air conditioners, the model is modified to account for both the heat and mass transfer of the cooling and dehumidifying coil (Threlkeld 1970; Brandemuehl et al. 1993). The model also assumes that the refrigerant temperature is constant at the SST through the
21、coil. The vapor compression system component models describe the steady-state performance of the system. Useful energy calculation models of residential HVAC equipment are obtained by modifiing the steady-state models to account for part-load cycling, the effects of fan power and heat, and in the ca
22、se of heat pumps, defrosting and crank-case heaters. The results of the model include the average power consumption of the compressor, indoor and outdoor fans, and ancillary equip- ment, as well as the temperature and humidity of the air deliv- ered to the building and the average run-time fraction
23、of the system. DATA REQUIREMENTS AND AVAILABILITY All energy calculation models for residential HVAC equipment rely on published data to describe the performance of the equipment being modeled. Some models require a map of performance over a range of psychrometric conditions and airflow rates to obt
24、ain regression curves. Since the component model uses first-principle relationships, it requires less infor- mation to identie the model parameters. In fact, given the compressor rating data, the steady-state model parameters can be identified using performance data at a single system oper- ating co
25、ndition. Table 1 shows the data required at this one rating point. AR1 Published Ratings The testing and reporting of data for air conditioners and heat pumps are defined by AM Standard 210/240-2003, Unitary Air-conditioning and Air-Source Heat Pump Equip- ment (AR1 2003). The purpose of the standar
26、d includes the establishment of test conditions, rating conditions, and mini- mum data requirements for published ratings to serve as guid- ance for the industry. More than 90% of the central air conditioners and air-source heat pumps produced in North America are rated according to the standard. Wh
27、ile the AR1 standard identifies test conditions and performance data calcu- lation methods, the tests are governed by ANSI/ASHRAE Stan- dard 37-1988, Methods of Testing for Rating Unitary Air- Conditioning and Heat Pump Equipment (ASHRAE 1988). The combination of standard testing methods and rating
28、requirements provides a framework for obtaining consistent and comparable performance results. The published data are publicly available at the AR1 Web site (AR1 2005). Table 1. Performance Data Required for Model Parameter Identification Air Conditioner Heat Pump Total cooling capacity (gross) Sens
29、ible cooling capacity (gross) Indoor coil airflow Indoor coil entering temperature Heating capacity (gross) Indoor coil airflow Indoor coil entering temperature Outdoor coil airflow Outdoor coil entering temperature Indoor coil entering humidity Outdoor coil airflow Compressor power Outdoor coil ent
30、ering temperature Indoor fan power Compressor power Outdoor fan power Indoor fan power Ancillary power Oiitdnnr fan nower ASHRAE Transactions: Research 21 5 ARI Standard 2 10/240-2003 specifies the standard test conditions for the performance ratings. Table 2 summarizes the tests and rating conditio
31、ns for cooling operation. The exact number of tests required by the standard varies with system characteristics, and several of the tests are performed to demonstrate robust and fault-free operation. The tests include both steady-state and cyclic operation with outdoor dry-bulb temperatures from 67F
32、 to 115F (19.4”C to 46.1”C) and indoor wet-bulb temperatures from 57F to 75F (13.9”C to 23.9”C). Table 3 gives a similar set of tests and rating condi- tions for air-source heat pumps. Both Tables 2 and 3 apply to single-stage equipment; equipment with multiple-speed compressors, multiple compressor
33、s, or compressor unloading capabilities have slightly different rating test conditions. In addition to the standard rating conditions, the ARI stan- dard specifies methods of calculating metrics by which to compare expected annual energy performance of an air condi- tioner or heat pump. The standard
34、 specifies the calculation of the cooling seasonal energy efficiency ratio (SEER) of an air conditioner based on the results oftests B, C, and D in Table 2. While test A is meant to represent Spica1 design conditions, test B is meant to represent more typical operating conditions in North America. T
35、ests C and D are conducted at the same conditions as test B and are meant to characterize the cyclic performance of the equipment, reflecting operation under typical conditions. The SEER is defined in the standard as the ratio of total cooling energy output to the total electric energy Table 2. AR1
36、Standard 2101240 Cooling Test Conditions Indoor Outdoor Unit Unit Dry-Bulb/ Dry-Bulb/ Wet-Bulb Wet-Bulb FIOF FIOF (OCIOC) (OCIOC) “A” cooling steady state 80.0167.0 95.0/75.0 (Standard rating conditions) (26.749.4) (35.0/23.9) “B cooling steady state 80.0/61.0 82.0/65.0 (26.7/19.4) (27.U18.3) “C” co
37、oling steady-state dry coil 80.0/57.0 82.0165.0 (26.7/13.9) (273118.3) “D” cooling cyclic dry coil 80.0/5 7. O 82.0/65.0 (26.7/13.9) (27.818.3) Low-temperature operation 67.0157.0 67.0157.0 (19.4/13.9) (19.4113.9) Insulation efficiency 80.0/75.0 80.0/75.0 (26.7/23.9) (26.7/23.9) Condensate disposal
38、80.0175.0 80.0r75.0 (26.7/23.9) (26.7/23.9) Maximum operating conditions 80.0/67.0 115.0/75.0 (26.7A9.4) (46.1/23.9) input during the normal usage period for cooling. For single- stage equipment, the standard specifies its calculation by SEER = (1 - 0.5 CD)EER, (3) where EER, is the ratio of the net
39、 total capacity to the total power input for test B (with units of Btu/Wh) and CD is the cyclic degradation coefficient as determined from tests C and D. However, the standard allows manufacturers to omit tests C and D, offering the use of a default CD = 0.25 in lieu of a measured valued obtained by
40、 tests C and D. For heat pumps, the standard defines a method for calcu- lating the heating seasonal performance factor (HSPF) as a metric to compare expected annual heating energy consump- tion. The method uses test results from the high- and low- temperature heating tests and the cyclic high-tempe
41、rature heating test, together with an assumed building load profile and frequency of occurrence of outdoor temperatures, to calculate the HSPF. As with the cooling SEER, the manufac- turers are offered the option to use a default cyclic degradation coefficient in lieu of a measured valued obtained b
42、y tests. It might be expected that the suite of tests required under AR1 Standard 210/240 offers a consistent and rich source of data with which to simulate the performance of an air condi- tioner or heat pump. Almost all of the data required for energy modeling, identified in Table 1, are available
43、 at all test condi- tions. Unfortunately, AR1 Standard 2 10/240 identifies mini- mum data reporting requirements for published ratings that are far more limited. For air conditioners, the standard only requires publication of total net cooling capacity from test A and the calculated SEER. For heat p
44、umps, the standard also requires publication of the net heating capacity at the high- temperature heating conditions and the calculated HSPF. The Table 3. AR1 Standard 210/240 Heating Test Conditions Indoor Outdoor Unit Unit Dry-Bulb/ Dry-Bulb/ Wet-Bulb Wet-Bulb OFfOF OFIOF (OCIOC) (OCIOC) High-temp
45、erature heating steady 70.0/60.0 47.0/43.0 state (Standard rating conditions) (21.1h5.6) (8.3/6.1) High-temperature heating cyclic 70.0160.0 47.0/43.0 (2 1.M 5.6) (8.316.1) Low-temperature heating 70.0/60.0 17.0/15.0 (2 1.M5.6) (-8.3/-9.4) Frost accumulation 10.0/60.0 35.0/33.0 (21.u15.6) (1.7/0.6)
46、Maximum operating conditions 80.0/- 75.0/65.0 (26.71- (23.9/18.3) 216 ASHRAE Transactions: Research online AR1 directory also includes the low-temperature heat- ing capacity in addition to these four required items. In most cases, the ARI performance ratings are also published in the technical liter
47、ature provided by the manufac- turers. Usually these ratings appear in the product data as tables of all the different combinations of outdoor units and matching indoor coils or air handlers for a particular model numberThe tables often include additional performance data besides the minimum data, s
48、uch as the rated power, EER, sensible capacity, or rated indoor and outdoor fan airflow rates. However, this varies from manufacturer to manufacturer and even from model number to model number for each partic- ular manufacturer. Table 4 shows the data that are typically available in the published li
49、terature for the top seven manu- facturers of residential air conditioners and heat pumps, repre- senting 99% of the North American market (DOE 2004). Manufacturer names are hidden to protect their identities. Note that no manufacturer typically provides all data required by the component model. Manufacturers Application Data In addition to the performance data obtained under ARI Standard 2 10/240, many manufacturers publish detailed heat- ing and cooling performance at other operating conditions. Depending on the manufacturer, it is possible to obtain compressor po
