Purdue University Purdue e-Pubs International and School of Conference

2014 The eP rformance Matching of Inverter Room Air Conditioner Bo Huang SHANG HITACHI, China, People's Republic of, [email protected]

Xiyuan Zhu SHANG HITACHI, China, People's Republic of, [email protected]

Hong Tao SHANG HITACHI, China, People's Republic of, [email protected]

Da Shi SHANG HITACHI, China, People's Republic of, [email protected]

Follow this and additional works at: http://docs.lib.purdue.edu/iracc

Huang, Bo; Zhu, Xiyuan; Tao, Hong; and Shi, Da, "The eP rformance Matching of Inverter Room Air Conditioner" (2014). International Refrigeration and Air Conditioning Conference. Paper 1449. http://docs.lib.purdue.edu/iracc/1449

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information. Complete proceedings may be acquired in print and on CD-ROM directly from the Ray W. Herrick Laboratories at https://engineering.purdue.edu/ Herrick/Events/orderlit.html 2343, Page 1

The Performance Matching of Inverter Room Air Conditioner

Bo HUANG1*; Xiyuan ZHU2; Hong TAO3; Da SHI4 1Shanghai Hitachi Electrical Appliances Co.Ltd, Technology Research Dept. Shanghai, China Contact Information (+8621-13701682356, [email protected])

2Shanghai Hitachi Electrical Appliances Co.Ltd, Technology Research Dept. Shanghai, China Contact Information (+8621-18321805082, [email protected])

3Shanghai Hitachi Electrical Appliances Co.Ltd, Technology Research Dept. Shanghai, China Contact Information (+8621-13501728323,[email protected])

4Shanghai Hitachi Electrical Appliances Co.Ltd, Technology Research Dept. Shanghai, China Contact Information (+8621-13564483789, [email protected])

ABSTRACT

Some key parameters like system charge amount, electronic expansion valve, frequency, indoor and outdoor air-flo w rate have been optimized using simulation tools for each matching condition of an inverter room air conditioner in the process of its performance matching. Thus, with the simulation tools and a new work process, the experiments of the performance matching of inverter room conditioner become more efficient, which will greatly shorten the test time, reduce workload and cut the cost.

1. INTRODUCTION

With the improvement of global energy conservation and environmental protection requirements, inverter room air conditioner is becoming increasingly attractive in the market thanks to its high energy saving ratio and flexibility in contrast to traditional constant speed air conditioner. Along with the growing market, many countries and organizations have announced new rules to standardize the performance of air conditioner. Ch ina issued its new APF standards for inverter room a ir conditioner, wh ich completes the comprehensive evaluation of the performance of a roo m air conditioner by investigating five different test points.

1.1 The Definition and Difficulties of Air Conditioner`s Performance Matching The performance matching of a ir conditioner is to ensure that the system can achieve its best performance under each assessment condition. Under the optimized condition, the system not only can meet the demand of cooling or heating capacity of every condition, but it should also operate with the highest COP value. Table 1: Comparison of Old and New Standards of China Item Old Standards New Standards Air conditioner type Fixe d -speed Various-speed 1) Intermediate Cooling Conditions 2) Rated Cooling Conditions Assessment points Rated Cooling Conditions 3) Intermediate Heating Conditions 4) Rated Heating Conditions 5) Low Temperature Heating Conditions 1) Refrigerant charge amount 2) Frequency of Compressor Adjustable parameters Refrigerant charge amount 3) Electronic expansion valve 4) Indoor air-flow rate 5) Outdoor air-flow rate Original Matching method Air- tests Air-enthalpy tests

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 2

In the performance matching of f i xe d -speed air conditioner, the traditional approach is to use air-enthalpy test matching method. In the case of an inverter room air conditioner with electronic expansion valve, the system performance is supposed to be evaluated under five different conditions which represent the most probable scenarios. The five conditions include intermediate cooling, rated cooling, maximum cooling, intermediate heating and rated heating. Each condition is described by several parameters such as refrigerant charge amount, compressor operating frequency, indoor and outdoor air-flow rate, etc. with the opening of electronic expansion valve directly related to the compressor suction superheat. For instance, each test lasts 2.5 hours, and total 20 tests (actually more) will spend 40 hours, if 8 hours a day, then the total matching test will require about one week, consuming a lot of human resources, experimental resources, and money, etc. Not to mention some other experiment breakdowns. In this case, both the considerable time consumption and huge investment makes the original air-enthalpy test method unattractive.

1.2 The improvement of the Test Matching Method In fact, multivariable parameters optimization needs a good mathematical log ic, and it is essential and complex to control an array of variable factors. But in the actual air-enthalpy matching experiments, single factor variable control method can ma ke laboratory wo rk faster and efficient. In this method, under each condition, only one variable parameter is investigated and optimized. Consequently, such method failed to take the inter-related influence among several variable parameters on the system at the same time . Th is paper selects a compromise approach between the actual test methods and mathematical methods. But this approach is rea lly constructive and effective for the performance matching of inverter room air conditioner especially for a company. So me processes and a comparison between this new approach and the traditional air-enthalpy tests are shown in Ta ble 2, Table 2: Comparison of the Air-enthalpy test method and new method

Process of work Old method (only air-enthalpy tests) New method (air-enthalpy test & simulation tools) 1. Test 1. Four diagnostic groups of air-enthalpy tests 2. Test 2. Calibration according to the four tests Blindness and require too much 3. Test 3. Calculation according to the four calibrated points time and money. 4. Test 4. Recommend results ``````` 4. Final air-enthalpy test validation

Detail descriptions about the new method are as follow: Step 1: do four different air-enthalpy tests, whose parameters do not need too much adjustment; S tep 2 : calibrate the simulation system with the results of the four air-enthalpy tests, the calibrated simulation accuracy must be maintained at less than 4%; Step 3: accomplish some calculations in a small-scale range of every parameter blew in the four conditions: intermediate cooling, rated cooling, intermediate heating and rated heating: 1) Refrigerant charge amount 2) Frequency of compressor 3) Suction Superheat 4) Indoor air-flow rate 5) Outdoor air-flow rate S tep 4 : recommend value or range of the parameters according to the calculation results S tep 5 : final test validation and slight adjustment.

Actually, this new method is based on the actual a ir-enthalpy tests and a combination of the a ir-enthalpy test method and simulation approach. Firstly, four diagnostic groups of air-enthalpy tes ts are performed and their parameters do not need too much adjustment. Secondly, s ome accurate ca librat ions to the simu lation system are performed just according to the four groups of air-enthalpy tests. After calibration, the simulation accuracy is maintained at less than 4%. Then accomplish some calculations in the vicinity of a small-scale range of the calibrated points. At last, the data is summarized and analyzed partially and globally. Thus every optimal range of the critical parameters is determined and experimenters can quickly complete these matching tasks according to the provided value or range of these parameters. So with the help of simulation tools, the experimental effort and cost can be greatly reduced, which will is definitely economical attractive to a company.

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 3

2. SIMULATION AND CALIBRATION OF THE COMPONENTS

2.1 Components Model and Algorithm of Air conditioner A basic air conditioner consists of four main components: compressor, indoor , outdoor heat exchanger, and throttling device. In simulation tools, each component has a specific model and its own algorith ms. When they are combined into a roo m air conditioning system, a comprehensive method to solve the system is required. This paper us es a mathematics model for the simulation of finned tube heat exchanger based on distributed parameter. The inverter compressor model is a s e mi-empirical model and electronic expansion valve model is also a s emi-empirical model. The simulation is solved by solving multiple equations simultaneously with numerical solvers. In this paper, an inverter room air conditioner (Rated Cooling Capacity: 3500W, Rated heating Capacity: 4200W) was analyzed and optimized.

2.2 Original Enthalpy Experiment Data All fo llo wing tests were done at the a ir-enthalpy experiment laboratory of Shanghai Hitachi Electrical Appliances Co. Ltd. The performance of the room air conditioner was measured according to certain experimental standard, which is specified in GB21455-2013 of China, and Ta ble 3 shows the main experimental conditions. Table 3: Enthalpy Experiment Conditions

Indoor and outdoor unit \ Indoor Indoor dry bulb temperature: Tdb Indoor wet bulb temperature: conditions （℃） Twb（℃） Outdoor(cooling/heating Test） 35/20 24/15 Indoor(cooling/heating Test） 27/7 19/6

Rated test means the cooling (or heating) capacity of the air conditioner reaches the target value on the nameplate; Intermediate test means the cooling (or heating) capacity of the air conditioner reaches half of the standard value on the nameplate. More information about experiment standard, please reference GB 21455-2013 of China. The original a ir-enthalpy test results are show in Table 4. All temperatures were measured using platinum resistances, whose measurement erro r is less than 0.2 ℃; other tested results like power, air-flow rate, capacity, pressure and so on, were all reliable with their relative errors within 2%.

Table 4: Original Enthalpy Experiment Data

Refrigerant R410A Rated capacity Rated Cooling Capacity:3500W; Rated heating Capacity:4200W Tested unit Inter-cooling Rated-cooling Inter-heating Rated-heating Frequency Hz 30 70 39 79 Tw-air ℃ 17.76 14.28 31.61 43.02 Td-air ℃ 15.82 13.03 19.1 22.62 3 Qvindoor m /h 614 591 621 614 3 Qvoutdoor m /h 1600 1600 1600 1600 Qc(Qh) W 1968 3442 2373 4484 Qsen W 1984 2633 2373 4484 Power W 460 1159 542 1466 EER/COP 4.28 2.97 4.37 3.06

Teva_in ℃ 16.3 12.61 27.76 35.64

Teva_out ℃ 14.3 7.47 47.54 80.11

Teva_mi d ℃ 15.4 10.87 27.94 35.71 Ts ℃ 19.5 11.91 3.84 1.16

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 4

Td ℃ 58 75.56 51.34 81.89

Tcond_in ℃ 57.6 74.77 2.79 -0.29

Tcond_mi d ℃ 42.1 47.99 3.21 1.07

Tcond_out ℃ 36 37.29 3.73 3.69 Pd[abs] MPa 2.54 2.989 2.273 3.401 Ps[abs] MPa 1.19 0.9 0.85 0.757

θ /500 405/500 170/500 105/500 140/500 M g 930 930 930 930

Air

T he wall of copper tube Refrigerant Temperature sensor

Figure 1 Schematic diagram of temperature sensor measurement point Since the measured temperature and the real temperature of the refrigerant pipe are not the same, temperature sensors are generally affixed to the outer wall of copper, shown in Figure 1, the temperatures measured are actually the combined effects of the refrigerant temperature in the copper tube and the air temperature of the outer copper tube. But as long as the calibration accuracy meets the requirement (within 5%), the accuracy of the calibration can be considered reliable in the simulation.

2.3 Calibration and Accuracy of the Simulation System In this simulation of air conditioner, simulat ion will also bring some deviations into the calculations. So there will be some errors between the simulation results and the air-conditioner test results. In order to ensure the accuracy and reliability of the simulation, every simulation model must be calibrated to reduce the calculation errors on the basis of the initial tested data in Table 5. Tabl e 5: Accuracy of simulation ABS ABS Location P_test P_sim T_test T_sim Test Residual Residual Unit [MPa] [MPa] [MPa] [℃] [℃] [℃] Compressor suction 0.902 0.900 0.000 11.91 11.51 -0.4 Rated Compressor outlet 2.989 2.989 0.000 75.56 83.97 8.41

Cooling Condenser outlet 2.989 2.877 -0.122 37.29 40.72 3.43 Condenser inlet 0.902 1.062 0.160 12.61 9.22 -2.39 Compressor suction 0.757 0.757 0.00 1.16 -1.53 3.46 Rated Compressor outlet 3.401 3.401 0.00 81.89 91.36 9.47

Heating Condenser outlet 3.401 3.382 -0.091 35.64 37.75 2.11 Condenser inlet 0.757 0.792 0.035 3.69 3.45 -0.24

In Table 5, the pressure values of across the condenser were not obtained from the experiment, but they were predicted according to the pressure at the two ends of the compressor. Although there was slight deviation between the simu lation ca librat ion results and the experimental data, it did not affect the overall system simulation accuracy. The simulation of other conditions are all based on the above four calibrations, fro m the origina l four points to mu ltip le points, and then gradually extend to all frequencies and all conditions. At the same time, So me experimental verifications have been made and as a result, the error between tested and calculated is less than 5%, which further indicates that the simulation is reliable. After the completion of the ca lib ration for a ll tested points and fulfilling the simulation requirement, calculation can continue in the vicinity of the experimental point calibration.

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 5

3. SINGLE VARIABLE LOCAL OPTIMIZATION OF THE PARAMETERS

As discussed above, several parameters affect the performance of inverter room air conditioner. The New APF standard demands the performance of room air conditioner as high as possible even in every test condition, which means each parameter should be set to an optimum value (or within the optimal range) to ma ke the a ir conditioner achieve overall excellent performance. As a result, the optimization of each factor on the air conditioner system performance becomes very important. The following discusses a single variable local optimization process of this room air conditioner.

3.1 Local Optimization of Re frige rant Charge Amount In general, the refrigerant charge amount of room air conditioner is the first important impact factor on the performance of the system, and our calculations range from 800g to 1040g. The figures are also with s ome change of compressor frequency, which will be further discussed in other chapter. 2160 4.60 Qc-30Hz 1980 Qc-27Hz 4.50 1800 4.40

1620 4.30

COPCOP-27Hz

1440 COP-30Hz 4.20

1260 4.10 COP Qc, P [W] 1080 Current Point 4.00 900 3.90

720 Power-30Hz 3.80 Power-27Hz 540 3.70

360 3.60 780 800 820 840 860 880 900 920 940 960 980 1000 1020 1040 Re f r i ge r a n t C h ar g e A m o u nt [ g ]

Figure 2 Performance vs. Ref charge amount (inter-cooling) 4000 3.05 3750 Qc-72Hz 3.03 Qc-70Hz 3500 3.00 3250 2.98

3000 2.95

2750 2.93

2500 COP-70Hz 2.90 COP COP-72Hz

Qc, P [W] 2250 2.88 2000 2.85 1750 Current Point 2.83 1500 Power-72Hz 2.80 1250 PowerPower-70Hz 2.78 1000 2.75 780 800 820 840 860 880 900 920 940 960 980 1000 1020 Re f r i ge r a n t C h ar g e A m ou n t [ g ]

Figure 3 Performance vs. Ref charge amount (rated-cooling)

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 6

2500 4.50 2250 4.47 Qh-3 6Hz 2000 Qh-3 9Hz 4.44 1750 4.40

1500 4.37

1250 4.34 COP-36Hz COP

1000 COP-39Hz 4.31

Qh, P [W] 750 Current Point 4.28 500 4.24 P o wer-39Hz 250 P o wer36Hz 4.21 0 4.18 780 800 820 840 860 880 900 920 940 960 980 1000 1020 1040 Re f r i g e r a n t C h a r g e A m o u nt [ g ]

Figure 4 Performance vs. Ref charge (inter-heating) 5000 3.18 4750 3.17 4500 3.16 4250 Qh-7 9Hz 3.15 Qh-7 7Hz 4000 COP-77Hz 3.14

3750 COP-79Hz 3.13

3500 3.12 3250 3.11

3000 3.10 COP 2750 3.09 Qh, P [W] 2500 3.08 2250 Current Point 3.07 2000 3.06 P o wer-79Hz 1750 3.05 P o wer-77Hz 1500 3.04 1250 3.03 1000 3.02 780 800 820 840 860 880 900 920 940 960 980 1000 1020 1040

Refrigerant Charge Amount [g] Figure 5 Performance vs. Ref charge (rated-heating)

Qc-30Hz means the cooling capacity of system when the compressor is running at 30Hz; Qh-79Hz means the heating capacity of system when the compressor is running at 79Hz. The meanings of other labels are in the same way.

Refrigerant charge amount is an integral factor. As we can see from Figure 2 to Figure 5, the optimum values for each air-enthalpy tes t conditions are not the same, but on a comprehensive comparison, we can see that the intermediate cooling conditions is more sensitive to refrigerant charge amount compared with other conditions, so refrigerant charge amount can be determined f ro m the intermediate cooling conditions. To this experiment, 930g is just close to the optimal value, so there is no need to do other optimization about refrigerant charge amount any more. In addition, fro m the Figure 2 to Figure 5, it also suggests that, when the operation frequency of the compressor is increased, the cooling (heating) capability Qc (or Qh) will increase, as well as system power. But on the contrary, the COP of the system will reduce, which will be particularly discussed in the next chapter.

3.2 Local Optimization of Compressor`s Running Frequency In the air conditioner system, the operating frequency of co mpressor is the second important factor that influencing the performance of air conditioner, which directly determines the cooling (heating) capacity and system power. Thus the optimization of the operating frequency of the compressor is particularly important. Figure 6 and Figure 7 showcase the frequency influence on the system performance under cooling and heating operation respectively.

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 7

5000 4.50 C O P ↑ 1 . 5 % 4500 Qcooling 4.30 Power 4000 4.10 COP b

3500 3500W 3.90

3000 3.70 2500 3.50 COP Rat ed Cooling 2000 a 3.30 1750W

Qcooling, P [W] P Qcooling, 1500 3.10 1000 Intermediate Cooling 2.90 27Hz 30Hz 70Hz 72Hz 500 2.70 0 2.50 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 fr e q u en cy [ H z ]

Figure 6 Performance vs. Frequency (Cooling cycle)

5000 5.00

4500 C O P ↑ 2 % 4.75 4350W 4000 d 4.50

3500 Qheating 4.25

Power 3000 4.00 COP

2500 3.75 COP [W] P ing, c Rated Heat ing 2175W eat 2000 3.50 h

Q 1500 C O P ↑ 1.6% 3.25 Intermediate Heating 1000 36Hz 39Hz 3.00 77Hz 79Hz 500 2.75 0 2.50 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 fr e q u en cy [ H z ]

Figure 7 Performance vs. Frequency (Heating cycle)

Cooling and heat capacity and COP are determined to represent the performance as shown in the Figure 6 and Figure 7, a conclusion can be drawn from the two figures: when the compressor frequency rises, the cooling (heating) capacity and power will increases, but the COP of the system has a best compressor frequency range. Additionally, the optimal frequency range is different between cooling and heating conditions. Actually, both of its intermediate conditions and rated conditions are at the right side of the curve peak, the COP of the air conditioner will decrease with the increase of compressor frequency, so the smaller the cooling (heating) capacity, the higher the COP of system. Furthermore, the cooling and heating capacity mus t meet the requirements of the test, as for this air conditioner, as long as the rated cooling capacity Qc is above 3500W, the intermediate cooling capacity Qc1 is above 1750W, the intermediate heating capacity Qh1 is above 2175W, and the rated heating capacity Qh is above 4350W (if no enough, it will need to increase), it is qualified for the system simulation. So the new suggested frequency is shown in origin color line in the figures.

Table 6 Compressor operating frequency optimization: Conditions Original frequency Suggested frequency COP effect Intermediate cooling 30Hz 27Hz Up 1.5% Rated cooling 70Hz 72Hz Up 1.6% Intermediate heating 39Hz 36Hz Up 2.0% Rated Heating 79Hz 77Hz Up 1.6%

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 8

3.3 Local Optimization of electronic expansion valve`s open rate Capillary or electronic expansion valve play a very important role in air conditioning system. This system uses electronic expansion valve, whose diameter is 0.9mm, and opening of the adjustable range is f ro m 0 to 500. The opening rate of the electronic expansion valve directly affects the compressor suction superheat and finally affects the overall performance of the system. The co mpressor suction superheat has a negative correlat ion relationship with the degree of the electronic expansion valve. The impact of superheat on the performance of the air conditioning system is show in Figure 8.

4.80 3800 4.60 3600 Qc-7 3Hz 4.40 3400 4.20 3200 COP-30Hz 4.00 3000

3.80

2800

3.60 OP

2600 C 3.40 Qc [W] 2400 3.20 2200 COP-73Hz 3.00 2000 Qc-3 0Hz 2.80 1800 2.60 1600 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 δ ( L o c a ti o n :E v a p o r a t o r O u tl e t ) Th [K ]

Figure 8 COP vs. Suction Superheat (Cooling Cycle)

There is no need to do other optimization about superheat, i.e . the opening rate of electronic expansion valve will need not too much adjustment according to Figure 8. Actually, other simulations and tests have been done in at least 5 different room air conditioners, and as a consequence, the bes t superheat range is 0 ~ 4K for cooling running to an air conditioner. While the superheat range of -1K ~ 2K is determined during heating conditions. Consequently, when the system is running in heating cycle, negative suction defined as small amount of liquid refrigerant at the suction will enhance the system performance. When suction superheat is negative, it means there is liquid refrigerant in the compressor suction. Other effects (such as “Liquid hammer”) will be not considered here, only the impact on system is focused. hh− suc 0 （ ） Superheat: ∆=Tsuperheat 1 Cp0

∆Tsuperheat , compressor suction superheat, Unit K;

Cp0 , bubble point Cp;

h0 , enthalpy of bubble point, unit kJ / kg;

hsuc , suction enthalpy of compressor, unit kJ / kg.

Under normal circumstances, hsuc > h0 , suction superheat is positive at this time, known as suction superheat;

When suction with liquid, hsuc < h0 , superheat is negative at this time, known as negative suction superheat. In these original tests, the ranges for superheat were accepted, namely the opening of the electronic expansion valve of every condition has already been well adjusted and does not need considerable adjust. In fact, in the four diagnostic groups of air-enthalpy tests, we have made some adjustments according to other simulat ion conclusion and get the results of other different room air conditioners, ranging from 1HP to 2HP.

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 9

3.4 Local Optimization of Indoor and Outdoor Air-Flow Rate In addition, indoor and outdoor air-flow rates also affect the system performance. When the system is running with a low compressor frequency, the total power of the system is greatly reduced, then the proportion of the indoor and outdoor `s power becomes lager. At the same time, like intermediate conditions, the provided air- flow rate may exceed the nominated rate, which will surely lead to e xt ra useless power. 4500 3.20 4250 Qh-7 9Hz 3.19 4000 Qh-7 7Hz 3.18 3750 COP-77Hz 3.17 COP-79Hz

3500 3.16 ] 3250 3.15 3000 3.14

2750 3.13 COP

Qh, P [W 2500 3.12 2250 3.11 2000 3.10 1750 P o wer-79Hz 3.09 P o wer-77Hz 1500 3.08 1250 Best Point Current Point 3.07 1000 3.06 950 1000 1100 1200 1300 1400 1500 1600 1700 Ou t d oo r A i r V o l um e [ m ³/ h ]

Figure 9 Performance vs. Qvindoor（rated-heating） 5000 3.60

4500 3.48

4000 Qh-7 9Hz 3.35

Qh-7 7Hz COP-77Hz 3500 COP-79Hz 3.23

3000 3.10 COP Qh, P [W] 2500 2.98 Best Point 2000 2.85 Current Point P o wer-779Hz9Hz 1500 P o wer-77Hz 2.73

1000 2.60 400 450 500 550 600 650 700 750 800 850 900 950 1000 1100 1200

Indoor Air Volume [m³/h] Figure 10 Performance vs. Qvoutdoor（rated-heating）

When in the rated heating condition, both of the existing indoor and outdoor a ir-flow rate have not reached the optimal value, indoor air-flow rate was too s mall ( Figure 9) . Therefore, the indoor air –flow rate can be further increased to 950m ³ / h, then system`s COP will significantly improve. While the outdoor air-flow rate was too big according to Figure 10, reduce outdoor air-flo w rate to 1300m ³ / h, will also benefit the increase of COP, but the effect is not so obvious. Of course, the demand of capacity mus t be met. But at the same time, changing indoor and outdoor air-flo w rate of rated conditions may generate new problems such as indoor noise, outdoor frost, if these two new issues are solved, the system performance will rise excellently. The results of intermediate heating are similar to those of rated heating, but there are slightly d ifferences shown in Figure 11 and Figure 13:

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 10

2500 4.60

2250 Qh-3 9Hz 4.48 2000 Qh-3 6Hz 4.36

1750 4.24 1500 4.12 COP-36Hz 1250 COP-39Hz 4.00 COP Qh, P [W] 1000 3.88 750 Best Point 3.76 500 3.64 P o wer-39Hz 250 Current Point P o wer-36Hz 3.52 0 3.40 400 500 600 700 800 900 1000 1100 1200 1300 1400 In do or A i r V o l um e [ m ³/ h ]

Figure 11 Performance vs. Qvindoor（inter-heating） 2500 4.80 2250 4.75 F Qh-3 9Hz 2000 ↑ 6.8% 4.70 C O P Qh-3 6Hz COP-36Hz

1750 COP-39Hz 4.65

1500 4.60

1250 Current Point 4.55 COP

Qh, P [W] 1000 New Point 4.50 D 750 P o wer-39Hz 4.45 P o wer--36Hz 500 4.40 250 A 4.35 0 4.30 950 1000 1100 1200 1300 1400 1500 1600 1700 Ou t d o or A i r V o lu m e [ m ³/ h ]

Figure 12 Performance vs. Qvoutdoor（inter-heating） 2000 4.50

1750 F QQcc-3 0Hz 4.46 Qc-2 7Hz C O P ↑ 4 . 4 % 1500 Current Point 4.42 COP-27Hz

COP-30Hz 1250 4.38

COP Best Point 1000 D 4.34 Qc, P [W]

750 4.30 P o wer-30Hz A 500 P o wer-27Hz 4.26

250 4.22 950 1000 1100 1200 1300 1400 1500 1600 1700 Ou t d o or A i r V o l u me [ m ³/ h ]

Figure 13 Performance vs. Qvoutdoor（inter-cooling）

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 11

In intermediate heating conditions, the original indoor air-flow rate 620 m³ / h is close to the optimal value ( Figure 10 ) ,there is no need to optimize anymore. While for the air –flow rate ( Figure 12 ) , if it is reduced from the original 1600 m³ / h down to 1200 m³ / h, COP will be significantly improved. Intermediate cooling conditions (Figure 13) and the intermediate heating is too much similar, indoor air has reached the vicinity of the optimum value, but its outer air-flo w rate is too big. Therefore when the outside air- flow rate is reduced from 1600 m ³ / h to 1200m ³ / h, the system COP will significantly increase . The following part explains the method to reduce the outside air-flow rate from from1600 m³ / h to 1200 m³ / h. By similarity theorem of fan, air-flow rate can be adjusted by adjusting the amount of wind speed of fan: Qn bb∝ （2） Qnaa P n bb∝ ()3 （3） Paan Q: the actual rate flow of fan, units: m³ / h n: the speed of fan, unit: rpm D: the diameter of fan impeller, unit: m So, if the air-flow rate decreases from 1600 m³ / h down to 1200 m³ / h, the fan speed is supposed to decrease to 75% of the original value, then the outdoor fan power will drop by about 50%.

3.5 Results and Discussion After the total process of this optimization, including the refrigerant charge amount, the electronic expansion valve (the compressor suction superheat), the running frequency of co mpressor, the indoor and outdoor air-flow rate, the overall performance of the compressor will be significantly improved. Fina l a ir-enthalpy tests were also carried on according to the optimal parameters. The results are pro mising, as shown in the following Ta ble 7 and Figure 14 and Figure 15:

Table 7: Optimization Program Summary Final Test Item Optimal Points Original Value New Optimal Value Predicted Optimal Results Validation Inter 1.Frequency 30Hz 27Hz Qc: 1968W→1806W Qc: 1812W cooling 2.Outdoor fan speed Down to the 75% of the original value COP: 3.06→3.15（↑4.66%） COP: 3.16 Rated Qc: 3442W→3553W (demand) Qc: 3526W 3. Frequency 70Hz 72Hz cooling COP: 2.97→2.93 COP: 2.94 Inter 4. Frequency 39Hz 36Hz Qh: 2373W→2170W Qh: 2153W heating 5. Outdoor fan speed Down to the 75% of the original value COP: 4.37→4.69（↑7.00%） COP: 4.70 Qh: 4484W→4392W Qh: 4401W Rated 6. Frequency 79Hz 77Hz heating COP: 3.06→3.15（↑2.96%） COP: 3.15

Cooling and heating capacity of basic requirements: Intermediate cooling capacity must be greater than 1750W; Rated cooling capacity must be greater than 3500W; Intermediate heating capacity must be greater than 2175W; Rated heating capacity must be greater than 4350W;

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 12

4500 Original Test 4250 Predicted Optimal Results 4000 Final Test Validation 3750 3500 3250 3000 2750 2500 2250 2000 Capacity (Qc Qh) or Capacity [W] 1750 1500 Inter cooling Rated cooling Inter heating Rated heating

Figure 14: Optimization Program Summary（Cooling or Heating Capacity）

4.80 4.69 4.7 4.60 4.40 4.37 4.20 Original Test 4.00 Predicted Optimal Results 3.80 Final Test Validation

COP 3.60 3.40 3.20 3.15 3.16 3.15 3.15 2.93 3.00 2.97 3.06 2.94 3.06 2.80 Inter cooling Rated cooling Inter heating Rated heating

Figure 15: Optimization Program Summary（COP）

The final results of test validation in the Figure 15 indicate that, although this simulation optimal method may be not rigorous enough in the logic of mathematic, but it is practically constructive. The calculations are accurate enough to make the optimization effect remarkable. This method is preferred in practice, the calculation results will significantly reduce the time and cost of the matching experiments. In fact, some other room air conditioner, their capacity range from 1HP to 1.5Hz, and 2HP, also have been optimized in this new “simulation & a ir-enthalpy test method”, the results are all decent enough for the performance matching of the inverter room air conditioners. In addition, the results are similar in general as summarized below. 1. Optimum refrigerant charge amount of the four test conditions are not the same, but the ultimate value can be preferentially determined by the intermediate conditions; 2. When the cooling and heating capacity demand is met, the lower the compressor frequency is, the higher the system COP will be, both in intermediate conditions and rated conditions; 3. In rated heating conditions, current indoor air-flo w rate increase and current outdoor air-flow rate decrease is both constructive to the performance of the system. But when the air-flo w rate changes, new problems like indoor noise and outdoor frost will need to be taken into consideration;

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 13

4. Under intermediate conditions, including intermediate cooling and intermediate heating, the original indoor air-flow rate is relatively reasonable, but the outdoor air-flo w rate (1600 m3 / h) is much larger than the optimal value (1200~1300 m3 / h). With 75% of the rated fan speed, the optimum condition can be achieved. 4. CONCLUSIONS With the simulat ion tools and the new work process, the experiments of the performance matching of inverter room conditioner become more e ffic ient. After optimization and final test validation, the COP of intermediate cooling increases by about 4.66%, intermediate heating COP increases by about 7.00%, and as well as rated heating rises about 2.96%. In general, this method will greatly shorten the test time, reduce the workload, and cut the cost.

NOMENCLATURE

Inter-cooling Intermediate Cooling Conditions Rated-cooling Rated Cooling Conditions Inter-heating Intermediate Heating Conditions Rated-heating Rated Heating Conditions f Frequency of compressor (Hz) Tw-air Indoor Web Bulb Temperature (℃) Td-air Indoor Dry Bulb Temperature (℃) 3 Qvindoor Indoor Air-flow Rate (m /h) 3 Qvoutdoor outdoor Air-flow R ate (m /h) Qc Cooling Capacity (W) Qh Heating Capacity (W) Qsen Sensible Capacity (W) Power Input Power of System (W) T Temperature (℃ Ps Suction Pressure (MPa) Pd Discharge Pressure (MPa) θ The Opening Ratio of Electronic Expansion Valve M Refrigerant charge amount (g) EER/COP Energy Efficiency Ratio/Coefficient of Performance

Subscripts eva_in Input of indoor heat exchanger eva_out Output of indoor heat exchanger eva_mid Middle of indoor heat exchanger cond_mid Middle of outdoor heat exchanger cond_in Input of outdoor heat exchanger cond_out Output of outdoor heat exchanger s Suction d Discharge

REFERENCES

Tao, H., Yang, J., Liu, C.H., Ye, K.J., Hu, C.G., Experimental investigations into rotary compressor performances with suction gas superheat, Journal of Refrigeration., vol.32, no.6: p.25-29. Daribi, A.E., Rice, C.K., 1981, A compressor simulation model with correlations for the level of suction gas superheat, ASHRAE Transactions., 187(3): 771-782. Wenhua Li, Simplified steady-state modeling for variable speed compressor, Carrier Corporation, Applied Thermal Engineering 50(2013)318-326. Liang, C.H., Zhang, X.S., Xu, G.Y., 2005, Simulative and Experimental Studies on the Effect of Superheat Degree on the Refrigeration System Performance, Fluid Machinery., vol.33, no.9: p.43-47.

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014 2343, Page 14

Han, L., Tao, L.R., Huang, L.H., Yang, Z.Q., Zhang, S.H., 2010, Experimental Study on the Effect of‘0’Superheat on the Inverter Rotary Compressor Refrigerant System Performance, Fluid Machinery., vol.38, no.10: p.53-57. Han, L., Tao, L.R., Zhen, Z.H., Yang, Z.Q., Chen, J., 2010, Experiment on Effect of Liquid-refrigerant Return on Performance of Refrigerant System with Rolling Rotor Compressor, Journal of Refrigeration., vol.31, no.4: p.22-25. Minimum allowable values of the energy efficiency and energy efficiency grades for variable speed room air conditioners. China National Institute of Standardization. GB21455-2013.

ACKNOWLEDGEMENTS

The authors would like to express gratitude to Xiaoyun JI for her help in completing this paper.

15th International Refrigeration and Air Conditioning Conference at Purdue, July 16-19, 2014