Study on the Performance of a New Ultra-Low Temperature Air Source Heat Pump (ASHP) Unit in Cold Regions

: Low air temperature and frosting have been reported as the critical factors that greatly attenuate the efﬁciency and performance of the ASHP in cold regions. In order to ensure the potential prevalence of the ASHP in cold regions of China, a new ultra-low temperature ASHP unit was developed, and the ﬁeld measurement was carried out in an ofﬁce building where these ASHP units were installed in Shanxi Province. Results showed that a coefﬁcient of performance ( COP ) of 1.83 was obtained at the ultra-low environmental temperature of − 25 ◦ C. Meanwhile, measured results indicated signiﬁcant frosting suppression and improved heating performance under three typical frosting conditions. In addition, long-term measurement results revealed that the mean COP and COP sys reached up to 3.34 and 2.63, respectively, indicating a higher performance in the cold regions of China. Consequently, the corresponding CO 2 emission reached 11.3 kg per year and per square meter, and the annual total cost on the unit reduced by 15.8% compared with the conventional ASHP, which meant that the total investment could be covered in the second year. The reduced CO 2 emission and the annual cost implied that the ASHP unit could produce better environmental and economic beneﬁts. Findings of this study revealed that this ultra-low temperature ASHP unit had a better performance under cold environment, which offered a possibility for the prevailing of the ASHP in cold or extremely cold regions, as well as could contribute to the carbon peaking and neutralization.


Introduction
As an effective energy-conservation technology, the Air Source Heat Pump (ASHP) has been gaining more and more attention throughout the world in recent years, and it has also been included as a renewable energy solution by the European Union, Japan, and China [1,2].The 'Clean Heating' policy of North China [3] has, since 2017, made the ASHP become an ideal option to get rid of coal.As a result, the ASHP has been regarded as a promising technology and is widely used in North China for heating.
The northwest of China is an extremely cold region; because of the large span between east and west, the climate of different regions is very different.According to the typical annual meteorological parameters in Northwest China, the lowest ambient temperature can reach −39 • C. Hence, the performance of the ASHP was inevitably compromised in the winter in Northwest China.The volumes of the refrigerant in the compressor expand due to the reduction of evaporative temperature and pressure, resulting the reduction of mass flow rate for the total refrigerant and the compromised heating capacity of the unit [4,5].It is well known that the lower the ambient temperature at which ASHP operates stably, the better the low temperature performance.Wu et al. [6] studied the system COP which ranged from 1.63 to 2.17 for an outdoor temperature range of −24.7 to 15.0 • C. Safa et al. [7] investigated ASHP performance at the Archetype Sustainable Twin House, Ontario, Canada; the COP of the ASHP ranged from 1.79 to 5.0 for an outdoor temperatures of −19 • C to 9 • C. Zhang et al. [8] studied the application results of an ASHP in Harbin, the coldest area in China, the COPs ranged from 1.04 to 2.44 for the low temperatures of −20.9 • C to −10.4 • C. Zhang et al. [9] found the COP of ASHP varied from 1.8 to 4.2 as the outdoor air temperature increased from −18.4 • C to 6.9 • C. In addition, the minimum ambient temperature of ASHPs in relevant Chinese Standards GB/T 25127.1 [10] and GB/T 25127.2[11] only reaches −20 • C, and the allowed COP cannot be lower than 1.8.It can be seen that the low temperature application ranges of ASHPs need to be further expanded, and the COP also needs to be further improved.
Moreover, the frosting would occur on the fin surface of the outdoor heat exchanger (the evaporator) of the ASHP unit in winter if the fin surface temperature of the outdoor heat exchanger was lower than the dew point temperature and the freezing temperature of the outdoor air [12][13][14][15].Previous studies showed that the mean COP of the ASHP units would be reduced by 35-60%, leading to a 30-57% reduction in heating capacity [16][17][18][19][20]. Therefore, it is significant to improve the adaptability of the ASHP in cold and extremely cold climates, as well as enhance the frosting suppression capacity, in order to ensure the high performance of the ASHP.
Considering the attenuated performance of the ASHP in Northwest China, a new ultra-low ASHP unit was developed and installed in an office building in Shanxi Province in China.And the field measurements were conducted in this building to study the performance of the new ASHP unit under typical frosting condition and long-term operation.Additionally, the environmental and economic benefits of the ASHP were also analyzed.The relevant results would provide a strong possibility for the prevailing of the ASHP in cold and extremely cold regions.

Description of the Ultra-Low Temperature ASHP Unit
A new ultra-low temperature ASHP unit was invented and developed by the authors, which can provide a reliable solution for the adaptability of the ASHP to cold and extremely cold climates, as well as enhance the frosting suppression at low environmental temperature.The schematic diagram of the new ultra-low ASHP unit was shown in Figure 1.A special rotary compressor that could replenish the refrigerant vapor and increase the enthalpy during the operation was adopted in the new ASHP unit.In addition, the refrigerating circle, specifically the system including the compressor, condenser and evaporator, was also improved and optimized.The detailed parameters of the components are given in Table 1.The test was carried out in an experimental psychrometric chamber, where required test conditions can be provided and maintained.The sensors and transducers were connected to a computerized measuring system to monitor in real-time and record the operating parameters.
As shown in Table 2, the performance of the new ASHP unit was tested at the dry bulb temperature of −25 • C, which was 5 • C lower than the test requirements on the performance for ASHP unit that recommended by the Chinese Standards of GB/T 25127.1 and GB/T 25127.2.Hence, referring the condition of −20 • C for ASHPs in the Chinese Standards of GB/T 25127.1 and GB/T 25127.2, the supply hot water temperature of the new ASHP unit was set at 41 • C, and COP was not lower than 1.8.In addition, based on ASHP products on the market, the heating power of the new ASHP unit was set at 7 kW at −25 • C conditions.Results showed that a heating power of 7.5 kW and a COP of 1.83 were observed, which were 7% and 1.7% higher than the expected design values.The validation indicated that the new ultra-low ASHP unit had a better heating capacity and efficiency under cold conditions.

Field Measurement Protocols
In order to test the practical performance of this new ASHP unit, an office building in Shanxi Province was selected, located in the cold region of China.In total, 2 ASHP units were installed to provide heating service for this building.A 38-day field measurement was carried out during the heating season in 2022 in this building.A sophisticated monitoring system was also available in this building to provide real-time data of these ASHP units.The temperature and humidity data of the region where the building is located was −14.7 • C and 47% in winter, respectively.The annual heating season lasts 151 days in total.The heating area of this building was 650 m 2 , and the radiant floor heating system was selected as the heating terminal.
The measuring system was established to acquire the critical data of the units, including the temperature and humidity sensors, the heating power sensors, and the electromagnetic flowmeter.These data were automatically recorded with a time interval of 60 s.For outdoor measurement, 1 recorder was installed near the outdoor heat exchanger to acquire the outdoor air temperature and relative humidity.The recorder had an accuracy of ±0.15 • C for temperature and ±3.5% for relative humidity.The measure ranges for the temperature and humidity were −20~70 • C and 0~100%, respectively.For indoor measurement, 1 recorder was installed in the testing room, with the same accuracies and ranges as those of the outdoor recorder.
For the measurement of ASHP units, 2 PT1000 sensors were installed in the suction pipe and discharge pipe of the compressor, respectively, to acquire the temperature data of the inlet and outlet of the compressor.In addition, 2 PT1000 temperature sensors were installed at the coil of the outdoor heat exchanger, one in the middle part and the other in the lower part of the coil, to monitor the temperature changes of the coils.These 4 PT1000 temperature sensors had an accuracy of ±0.15 • C and the range of −40~140 • C.Meanwhile, 2 pressure sensors were installed in the suction line and discharge line of the compressor to acquire the pressure changes of the compression during the whole field measurement.The accuracy of the pressure sensor was ±0.4% of the full range, and the pressure ranges were 0~25 bar and 0~40 bar, respectively.
For the measurement of the water, 2 PT1000 temperature sensors were attached to the supply and return pipes of the unit to acquire the changes of the supply and return water temperature.The accuracy and range were the same as that used in the ASHP unit measurement.In addition, an electromagnetic flowmeter was also installed at the supply water pipe of the unit to monitor the changes of flow rates of the circulating water.The accuracy of the flowmeter was ±0.5% of full range, and the measuring range was 0.5~10 m 3 /h.
Additionally, a digital camera with a resolution of 150 pix was installed near the outdoor heat exchanger to record the real-time frosting process on the fins.Three heating power sensors were also used in the measurement to acquire the power input of the ASHP units and the circulating water pumps.
Finally, in terms of the frosting map that was presented by Wang et al. [21], 3 typical frosting conditions were selected in the field measurement to scrutinize the frosting behaviors of the ultra-low temperature ASHP units.The details are given in Figure 2.Among these 3 typical frosting conditions, case 1, case 2, and case 3 of this study correspond to the severe frosting region, moderate frosting region, and mild frosting region, respectively.Moreover, the reverse-cycle defrosting method was used in the tested ASHP unit.For reverse-cycle defrosting, the outdoor heat exchanger changed from evaporator to condenser, while the indoor heat exchanger became the evaporator, and the energy originally used to heat a room would be used to heat the outdoor exchanger [22].Furthermore, a conventional defrosting initiation method, the Temperature-Time (T-T) method [1,21], was used to initiate defrosting of the two field ASHP.

Data Analysis
In this study, the COP was selected as a critical indictor for the frosting-suppression and heating performance of the ASHP unit, as usually used in similar studies.Moreover, the frosting time, the suction and discharge pressures, and the temperature of the coils were also recorded and analyzed.The COP indicates the heating capacity of the ASHP unit by consuming a certain amount of electric power, it can be calculated using Equations ( 1) and (2).
tively.Moreover, the reverse-cycle defrosting method was used in the tested ASHP unit.For reverse-cycle defrosting, the outdoor heat exchanger changed from evaporator to condenser, while the indoor heat exchanger became the evaporator, and the energy originally used to heat a room would be used to heat the outdoor exchanger [22].Furthermore, a conventional defrosting initiation method, the Temperature-Time (T-T) method [1,21], was used to initiate defrosting of the two field ASHP.

Data Analysis
In this study, the COP was selected as a critical indictor for the frosting-suppression and heating performance of the ASHP unit, as usually used in similar studies.Moreover, the frosting time, the suction and discharge pressures, and the temperature of the coils were also recorded and analyzed.The COP indicates the heating capacity of the ASHP unit by consuming a certain amount of electric power, it can be calculated using Equations ( 1) and (2).In the above formulas, q was the instant heating capacity of the unit and could be calculated using q = mwcpΔTw.Where mw was the flow rate of the water, cp was the specific heat at constant pressure, and ΔTw was the temperature differences between the supply and return water.W represented the power of the units, P represented the power input of the water pump, and tn denoted the defrosting time of the unit.
In addition, the equivalent energy consumption Qs, the annual cost AC, as well as the payback period of the additional investment ths were selected as the economic and environmental criteria of the unit.These indexes can be calculated using Equations ( 3)-( 6).
In the above formulas, q was the instant heating capacity of the unit and could be calculated using q = m w c p ∆T w .Where m w was the flow rate of the water, c p was the specific heat at constant pressure, and ∆T w was the temperature differences between the supply and return water.W represented the power of the units, P represented the power input of the water pump, and t n denoted the defrosting time of the unit.
In addition, the equivalent energy consumption Q s , the annual cost AC, as well as the payback period of the additional investment t hs were selected as the economic and environmental criteria of the unit.These indexes can be calculated using Equations ( 3)- (6).
In the calculation of the Q s , q represented the calorific value of the standard coal (29.307MJ/kgce), and Q h was the total heating capacity of the unit, η t denoted the thermal efficiency when using the conventional energy source, and the value was 0.7 when using the coal as the heating source [23].D was power input corresponding to the standard coal, and it was 0.3 kgce/kWh in terms of the latest data published by National Bureau of Statistics in China [24].COP sys represented the heating performance efficient of the heat pump system.C o was the annual operation cost of the unit.C m meant the annual maintenance cost, which was usually regarded as 6% of the total cost of equipment [24].(A/P,i,j) was the coefficient of the payback.C was the total initial investment of the ASHP unit, B meant the net salvage value of the ASHP unit.i denoted the constant discount rate of the unit, which was set as 4.594%, according to a previous study [25].j represented the whole life circle of the unit, ∆C was the additional investment, and ∆C o represented the annually saved cost.
Finally, the relative uncertainty analysis was used in the data analysis, considering that the errors due to the direct measurement would inevitably transfer to the indirect measured data and thus produce indirect measurement errors [26].The relative uncertainty U r was selected in this study to eliminate the indirect errors.The related calculation was given as Equations ( 7)- (9). ) where N represented the indirect measured parameter, such as COP and instant heating capacity.X, Y, and Z were the related parameters that were used to calculate N. δ N meant the standard uncertainty.

Environmental Data during the Measurement and the Predicted Frosting Levels
The outdoor environmental conditions during the 38-day field measurements were recorded and analyzed.Results were depicted in Figure 3.The daily air temperature ranged between −1.85 • C and 15 • C, and the daily relative humidity ranged between 24.4% and 81.3%.The daily outdoor air temperature and relative humidity were negatively correlated.The mean daily outdoor air temperature and relative humidity were 5.7 • C and 45.8%, respectively.Based on the outdoor temperature and relative humidity during the field measurements and the frosting map developed by Wang et al. [21], the frosting levels during the whole field measurement were predicted, and the results are shown in Figure 4.The severe frosting condition accounted for 7.8% of the whole operation time of the ASHP unit, while the moderate frosting and mild frosting accounted for 13.1% and 19.3%, respectively.The total frosting period could reach up to 40.3% of the whole 38 days.Additionally, as can be seen in Figure 4, the hourly minimum value of the outdoor ambient temperature reached −7 °C.Based on the outdoor temperature and relative humidity during the field measurements and the frosting map developed by Wang et al. [21], the frosting levels during the whole field measurement were predicted, and the results are shown in Figure 4.The severe frosting condition accounted for 7.8% of the whole operation time of the ASHP unit, while the moderate frosting and mild frosting accounted for 13.1% and 19.3%, respectively.The total frosting period could reach up to 40.3% of the whole 38 days.Additionally, as can be seen in Figure 4, the hourly minimum value of the outdoor ambient temperature reached −7 • C. ments and the frosting map developed by Wang et al. [21], the frosting levels during the whole field measurement were predicted, and the results are shown in Figure 4.The severe frosting condition accounted for 7.8% of the whole operation time of the ASHP unit, while the moderate frosting and mild frosting accounted for 13.1% and 19.3%, respectively.The total frosting period could reach up to 40.3% of the whole 38 days.Additionally, as can be seen in Figure 4, the hourly minimum value of the outdoor ambient temperature reached −7 °C.

Actual Performance of the New ASHP Units under three Typical Frosting Conditions
Based on the predicted frosting levels of the three cases, it was necessary to observe the actual frosting levels when the ultra-low temperature ASHP units was used during

Actual Performance of the New ASHP Units under Three Typical Frosting Conditions
Based on the predicted frosting levels of the three cases, it was necessary to observe the actual frosting levels when the ultra-low temperature ASHP units was used during the field measurement.The images of the frosting levels on the outdoor heat exchanger fins before the ASHP unit started to defrost are given in Table 3, as well as shown the recorded frosting time.The results indicated that only moderate frosting was observed on the outdoor fins of the ultra-low temperature ASHP units in case 1, under the severe frosting condition as predicted by the frosting map.Specifically, the frosting only occurred on the upper parts and the bottom of the outdoor heat exchangers, the frost did not cover all the fin surface.However, in case 2, predicted as the moderate frosting condition, only mild frosting was observed, and no frost was spotted in case 3, regarded as the mild frosting condition.These images suggested that the frosting of the ultra-low temperature ASHP units was greatly attenuated compared with the traditional ASHP unit.the field measurement.The images of the frosting levels on the outdoor heat exchanger fins before the ASHP unit started to defrost are given in Table 3, as well as shown the recorded frosting time.The results indicated that only moderate frosting was observed on the outdoor fins of the ultra-low temperature ASHP units in case 1, under the severe frosting condition as predicted by the frosting map.Specifically, the frosting only occurred on the upper parts and the bottom of the outdoor heat exchangers, the frost did not cover all the fin surface.However, in case 2, predicted as the moderate frosting condition, only mild frosting was observed, and no frost was spotted in case 3, regarded as the mild frosting condition.These images suggested that the frosting of the ultra-low temperature ASHP units was greatly attenuated compared with the traditional ASHP unit.Moreover, the frosting time recorded by the camera for these three cases was 109 min, 114 min, and 0 min (no frost), respectively.The frosting time of the new ASHP units was longer than that predicted by Reference [21], which produced strong evidence for the enhanced frosting-suppression capability of the ultra-low temperature ASHP unit.the field measurement.The images of the frosting levels on the outdoor heat exchanger fins before the ASHP unit started to defrost are given in Table 3, as well as shown the recorded frosting time.The results indicated that only moderate frosting was observed on the outdoor fins of the ultra-low temperature ASHP units in case 1, under the severe frosting condition as predicted by the frosting map.Specifically, the frosting only occurred on the upper parts and the bottom of the outdoor heat exchangers, the frost did not cover all the fin surface.However, in case 2, predicted as the moderate frosting condition, only mild frosting was observed, and no frost was spotted in case 3, regarded as the mild frosting condition.These images suggested that the frosting of the ultra-low temperature ASHP units was greatly attenuated compared with the traditional ASHP unit.Moreover, the frosting time recorded by the camera for these three cases was 109 min, 114 min, and 0 min (no frost), respectively.The frosting time of the new ASHP units was longer than that predicted by Reference [21], which produced strong evidence for the enhanced frosting-suppression capability of the ultra-low temperature ASHP unit.the field measurement.The images of the frosting levels on the outdoor heat exchanger fins before the ASHP unit started to defrost are given in Table 3, as well as shown the recorded frosting time.The results indicated that only moderate frosting was observed on the outdoor fins of the ultra-low temperature ASHP units in case 1, under the severe frosting condition as predicted by the frosting map.Specifically, the frosting only occurred on the upper parts and the bottom of the outdoor heat exchangers, the frost did not cover all the fin surface.However, in case 2, predicted as the moderate frosting condition, only mild frosting was observed, and no frost was spotted in case 3, regarded as the mild frosting condition.These images suggested that the frosting of the ultra-low temperature ASHP units was greatly attenuated compared with the traditional ASHP unit.Moreover, the frosting time recorded by the camera for these three cases was 109 min, 114 min, and 0 min (no frost), respectively.The frosting time of the new ASHP units was longer than that predicted by Reference [21], which produced strong evidence for the enhanced frosting-suppression capability of the ultra-low temperature ASHP unit.Moreover, the frosting time recorded by the camera for these three cases was 109 min, 114 min, and 0 min (no frost), respectively.The frosting time of the new ASHP units was longer than that predicted by Reference [21], which produced strong evidence for the enhanced frosting-suppression capability of the ultra-low temperature ASHP unit.
Furthermore, the heating performance of the new ASHP units were also scrutinized based on the measured data.The results are given in Figure 5.The outdoor air temperature and relative humidity were stable for all three cases.In case 1, the coil temperature of the outdoor heat exchanger of the new ASHP units was reduced, due to the effect of moderate frosting on the fin.The lowest coil temperature was found at −18.4 • C. Consequently, the heating capacity and the COP of the new ASHP units was reduced by 35.5% and 30.3% in case 1, respectively.However, both the heating capacity and the COP of the unit in case 2 and case 3 were found to be much more stable than case 1.The heating capacity and COP of case 2 were 14.1 kW and 3.1, respectively, while for case 3, they were 13.1 kW and 2.8, respectively.The results suggested that both case 2 and case 3 were less affected by the frosting.These findings were in accordance with the results in Table 3. Specifically, the moderate frosting attenuated the performance of the ASHP unit, while the mild frosting and zero frosting had less negative contribution to the performance of the unit.Furthermore, Table 4 shows the relative uncertainty values of COP for the measured unit in the three cases.The relative uncertainty values of COP in the three field measurement cases ranged from 3.6% to 4.3%, suggesting an acceptable measuring accuracy.In order to investigate the long-term performance of the ultra-low temperature ASHP unit, all these indexes used in field measurements were also recorded during the regular However, both the heating capacity and the COP of the unit in case 2 and case 3 were found to be much more stable than case 1.The heating capacity and COP of case 2 were 14.1 kW and 3.1, respectively, while for case 3, they were 13.1 kW and 2.8, respectively.The results suggested that both case 2 and case 3 were less affected by the frosting.These findings were in accordance with the results in Table 3. Specifically, the moderate frosting attenuated the performance of the ASHP unit, while the mild frosting and zero frosting had less negative contribution to the performance of the unit.Furthermore, Table 4 shows the relative uncertainty values of COP for the measured unit in the three cases.The relative uncertainty values of COP in the three field measurement cases ranged from 3.6% to 4.3%, suggesting an acceptable measuring accuracy.In order to investigate the long-term performance of the ultra-low temperature ASHP unit, all these indexes used in field measurements were also recorded during the regular heating season of the office building.The results are shown in Figure 6, including the indoor air temperature T n , the heating capacity per square meter q 0 , the temperature differences between the supply and return water ∆T, and the COP.
As depicted in Figure 6a, the mean indoor air temperature during the heating season reached up to 24.56 • C.And the mean temperature difference between the supply and return water was 4.18 • C, which suggested that the load rate reduced gradually as the outdoor air temperature increased, but the temperature differences between the supply and return water stayed stable, leading to a higher indoor air temperature of up to 28 • C in the last part of the whole measurement.In addition, it was found that the heating capacity per square meter q 0 reached 21.34 W/m 2 , and the mean COP was 3.34 during the whole measurement, which implied the improved continuous heating performance and frosting-suppression capability under cold environmental conditions.
Figure 6b describes the system performance of the ASHP unit.Changes of the indices, such as the heating capacity per square meter of the system (q sys ), the heating performance efficient of the heat pump system (COP sys ), and the power input of the water pump (P), were analyzed.During the heating season, the mean q sys reached up to 20.97 W/m 2 , the mean COP sys was 2.63, and the mean power input of the water pump was 1.0 kW.As shown in Table 5, by conducting the relative uncertainty analysis, the relative uncertainty of the calculated q 0 , COP, q sys , and COP sys were 3.7%, 4.3%, 3.8%, and 4.6%, respectively.These small uncertainty values revealed that the indirect errors were greatly eliminated, and the results calculated were reliable.
Buildings 2023, 13, x FOR PEER REVIEW 10 of 14 in the last part of the whole measurement.In addition, it was found that the heating capacity per square meter q0 reached 21.34 W/m 2 , and the mean COP was 3.34 during the whole measurement, which implied the improved continuous heating performance and frosting-suppression capability under cold environmental conditions.Figure 6b describes the system performance of the ASHP unit.Changes of the indices, such as the heating capacity per square meter of the system (qsys), the heating performance efficient of the heat pump system (COPsys), and the power input of the water pump (P), were analyzed.During the heating season, the mean qsys reached up to 20.97 W/m 2 , the mean COPsys was 2.63, and the mean power input of the water pump was 1.0 kW.As shown in Table 5, by conducting the relative uncertainty analysis, the relative uncertainty of the calculated q0, COP, qsys, and COPsys were 3.7%, 4.3%, 3.8%, and 4.6%, respectively.These small uncertainty values revealed that the indirect errors were greatly eliminated, and the results calculated were reliable.According to the long-term field measurement in the heating season, the total heating load was 174,821.8MJ per year.In terms of the calculation of the equivalent energy consumption Q s , the Q s of the ultra-low temperature ASHP unit in the heating season was 2961.2 kgce.
Based on the calculated Q s , the annual CO 2 emission reduction in the ASHP unit was also calculated through the standard coal coefficient method recommended by GB/T 50801-2013.Results showed that the carbon emission reduction in the ASHP was 7314.2 kg per year.Considering the heating area of the office building in this study, the carbon emission reduction in the new ASHP unit was 11.3 kg per year per square meter, which produced great environmental benefits compared with the traditional central heating system through the coal-fired boilers in thermal power plants.

Economic Benefits Analysis
The economic benefits were analyzed in terms of the total annual cost (AC), and the payback period of the additional investment (t hs ) on the ASHP unit.Less annual cost corresponds to the better choice.
According to the local power price 0.47 Yuan/kWh that was provided by the local government, and the mean COP of the ASHP (2.1) recommended by a report published by Chinese Academy of Building Science [27], the annual costs of operation of the conventional and the ultra-low temperature ASHP were 10,901 Yuan and 8687 Yuan, respectively.In addition, the initial investment C and net salvage B of the conventional ASHP unit were 6800 Yuan and 680 Yuan, respectively, while for the new ASHP unit, they were 9000 and 900 Yuan, respectively.Furthermore, as recommended by Reference [24], the annual cost of maintenance was 6% of the initial investment.Therefore, the cost on maintenance was 408 Yuan for the conventional ASHP, and 540 Yuan for the new ultra-low temperature ASHP unit.
Therefore, the total annual costs (AC) of the convectional and new ASHP unit were 11,914 and 10,027 Yuan, respectively.The detailed results were given in Table 6.The results suggested that the total annual cost of the ultra-low temperature ASHP unit was 1887 Yuan less than that of the conventional ASHP unit, corresponding to a cost reduction of 15.8%.The payback period of the additional investment (t hs ) on the ASHP unit is another important economic index.In order to realize the performance under cold environment, Additional investment was required to add on critical components, including the outdoor heat exchanger, the outdoor fan and the sheet metal processing.The detailed comparison of the cost on critical components between the conventional and ultra-low temperature ASHP unit was provided in Table 7.

Figure 1 .
Figure 1.Schematic diagram of the new ultra-low ASHP unit.

Figure 2 .
Figure 2. Typical frosting condition in the frosting map developed by Wang et al. [21].

Figure 2 .
Figure 2. Typical frosting condition in the frosting map developed by Wang et al. [21].

Figure 3 .
Figure 3. Changes of the daily outdoor temperature and relative humidity during the field measurements.

Figure 3 .
Figure 3. Changes of the daily outdoor temperature and relative humidity during the field measurements.

Figure 4 .
Figure 4. Distribution of the frosting levels during the whole field measurement.

Figure 4 .
Figure 4. Distribution of the frosting levels during the whole field measurement.

Figure 5 .
Figure 5. Heating performance of the ASHP under typical frosting conditions.

Figure 6 .
Figure 6.Performances of the new ASHP unit and system during long term operation.(a) Perform ances of the new ASHP unit during long term operation.(b) Performances of the new ASHP system during long term operation.

Figure 6 .
Figure 6.Performances of the new ASHP unit and system during long term operation.(a) Perform ances of the new ASHP unit during long term operation.(b) Performances of the new ASHP system during long term operation.

Table 1 .
Components parameters of new ASHP unit.

Table 1 .
Components parameters of new ASHP unit.

Table 2 .
Comparison between the designed and measured values of the new ASHP unit.

Table 3 .
The actual frosting levels and the recorded frosting time of the new ASHP.

Table 3 .
The actual frosting levels and the recorded frosting time of the new ASHP.

Table 3 .
The actual frosting levels and the recorded frosting time of the new ASHP.

Table 3 .
The actual frosting levels and the recorded frosting time of the new ASHP.

Table 4 .
The relative uncertainty values of COP for the test unit in the three field test cases.
3.3.Long-Term Performance of the New ASHP in Regular Heating Season

Table 4 .
The relative uncertainty values of COP for the test unit in the three field test cases.

Table 5 .
The relative uncertainty values of COP for both units in the three field test cases.ParameterUr(q0) Ur(COP) Ur(qsys) Ur(COPsys)

Table 5 .
The relative uncertainty values of COP for both units in the three field test cases.

Table 6 .
Comparison of the total annual cost of the conventional and ultra-low temperature ASHP.