ASHRAE AB-10-020-2010 Development of an Energy Meter Using a Pump Flow Station.pdf

1、2010 ASHRAE 569ABSTRACTHeating and cooling energy consumption measurements are critical for operations, controls, and fault detection and diagnosis (FDD) of heating, ventilation and air conditioning (HVAC) systems. Generally water flow has to be measured in order to determine energy consumption in e

2、ither chilled water systems or hot water systems. Cost-effective and accurate water flow measurements are essential to develop energy meters. Since pump performance relates the actual pump water flow to pump head and power, theoretically the water flow through a pump can be determined by other pump

3、performance characteristics, such as pump head and motor power. This paper presents the theoretical model of pump flow stations based on measured pump head and motor power, and the experiments and results of a cooling energy meter using a pump flow station developed on the chilled water system at a

4、facility. INTRODUCTIONHeating and cooling energy consumption measurements are critical for operations, controls, and fault detections of HVAC systems. In large commercial HVAC systems, the chilled water system may include multiple chillers and a ther-mal storage tank. Either the chillers or the stor

5、age tank will load and unload based on the system cooling load or consump-tion rate (McQuiston et al. 2005). The energy consumption is closely related to HVAC system operations. Fault or ineffi-cient operations will lead to energy consumption levels 15% - 30% greater than optimal (Friedman et al. 20

6、07). A simplified tool was developed to detect HVAC system faults by measur-ing heating and cooling energy consumption at the whole building level and calculating energy consumption deviation from the prediction (Friedman et al. 2007). Generally water flow has to be measured in order to deter-mine e

7、nergy consumption in either chilled water systems or hot water systems. The water flow measurements are critical to develop heating and cooling energy meters. The water flow measurement shares the same principle as the air flow measurement. The flow rate is often determined by measuring pressure dif

8、ference across an orifice, nozzle or venturi tube (ASHRAE 2001). However, valves, dampers, bends and fittings upstream from the flow devices can cause errors. Long, straight pipe or duct should be installed upstream and downstream the flow devices, to assure fully developed flow for the proper measu

9、rements. These conditions are hard to be satisfied in actual systems. Meanwhile, for existing systems, extremely high installation and retrofit cost may make it impossible to install a new flow meter. On the other hand, fans and pumps are always installed in HVAC systems. Since fan or pump performan

10、ce relates the actual flow rate to fan or pump head and power, theoretically the flow rate can be determined by other fan or pump perfor-mance characteristics, such as fan or pump head and motor power.To develop an accurate and reliable airflow measurement device, Liu (2002) proposed a fan airflow s

11、tation, which deter-mines airflow using measured fan speed and head associated with an in-situ fan head curve. Since this fan airflow station uses the fan head as a measured parameter, thereafter, we will call this fan airflow station as the fan-head-based fan airflow station. Experiments were condu

12、cted in a full size air handling unit (AHU) in a laboratory (Yuill et al. 2003). The direct measured airflow showed excellent agreement with the fan Development of an Energy Meter Using aPump Flow StationGang Wang, PhD, PE Mingsheng Liu, PhD, PE David E. Claridge, PhD, PEMember ASHRAE Member ASHRAE

13、Fellow ASHRAEGang Wang is an assistant professor in the Department of Civil and Architectural Engineering, Texas A and the second part from August 13 to August 14 was used to validate the pump flow station. The measured VFD frequency data shows that the pump speed was not stable with several sudden

14、and large fluctuations during the calibration. As we know, the basic equations, Equations. (1) and (8), are only satisfied under steady states. Flow accel-eration with a sudden increase of the pump speed will consume more power than under a steady state, while flow deceleration with a sudden decreas

15、e of the pump speed will consume less power than under a steady state. To eliminate the acceleration and deceleration impacts on the motor power, only data with pump frequency change less than 0.5Hz within 4 minutes will be remained to calibrate the motor and pump efficiency. The overall efficiency,

16、 the product of the pump and motor efficiency, can be calculated using these filtered data. (9)For each set of effective data, the overall efficiency and the ratio of the pump head to the pump flow square were calcu-lated. Figure 7 shows the overall efficiency data with the ratio of the pump head to

17、 the pump flow square with two different motor power ranges, 3.0-7.4 kW (gray marks) and 7.7-15.2 kW (black marks). It can be seen that the overall efficiency is a function of the ratio of the pump head to the pump flow square as well as the motor power.The motor efficiency can be estimated based on

18、 the motor manufacturer data. Normally the motor efficiency data at 25%, 50%, 75% and 100% of nominal motor power can be found from the manufacturer data sheet. It means that the motor effi-ciency may not be accurate enough with motor power less than 25%. Since the pump with low motor power will ope

19、rate at low water flow rates and the cooling energy consumption within low water flow rates has less impact on weekly energy usage, it is acceptable to use the manufacturer data to determine the motor efficiency function in this application. The motor installed on the chilled water pump is a 50HP, 1

20、800rpm, ODP (with an Open, Drip Proof enclosure) motor. Table 1 shows the motor efficiency data of this motor.The motor efficiency data in Table 1 were applied to determine constants, a and b, in Equation (5) using the least square method. The sum of the squares of the residuals is expressed as:(10)

21、The constant factors related to the motor efficiency were determined based on the data in Table 1.a = 0.08448b = 0.02845Since the nominal power of the 50HP motor is 37.2 kW, the motor efficiency can be expressed as a function of the measured motor power in kW.(11)Figure 5 Measured motor power and pu

22、mp head.overallHQWmotor-=S =motor i,Wmotor i,a motor i,Wmotor i,()2+ bWmotor i,+2i 1=4motorf2Wmotor() =1 14+ 0.08448 Wmotor37.2 0.02845+2 0.08448 Wmotor37.2-2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010,

23、 Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.574 ASHRAE TransactionsThen the pump efficiency can be derived from the calcu-lated overall efficiency and the c

24、alibrated motor efficiency. Figure 8 shows the pump efficiency with the ratio of the pump head to the pump water flow square. The pump efficiency curve in Figure 8 can be regressed as a second order polynomial expression of the ratio of the pump head to the pump water flow square.In IP units (H in P

25、si and Q in 1000GPM):(12a)In SI units (H in kPa and Q in m3/s): (12b)Now a pump flow station has been developed using Equa-tions (11) and (12). The chilled water can be calculated using the measured motor power and pump head. The measured data from August 13 to August 14 were used to validate the pu

26、mp flow station. Figure 9 compares the water flow measured by the ultrasonic water flow meter and the water flow calculated by the developed pump flow station. The gray line shows the Figure 6 Measured water flow and VFD frequency.Figure 7 Overall efficiency.pumpf1HQ2-=0.00000986415 HQ2()2 0.0036249

27、9 HQ2() 0.898484+pumpf1HQ2-=0.0000425848 HQ2()2 0.00753189 HQ2() 0.898484+2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, o

28、r transmission in either print or digital form is not permitted without ASHRAEs prior written permission.2010 ASHRAE 575measured water flow using the ultrasonic meter while the black line shows the calculated water flow using the pump flow station. Figure 10 shows normalized flow difference (%) betw

29、een the calculated flow and the measured flow. The normalized value is based on the maximum measured flow during the validation period, 744 GPM or 2.82 m3/s. The normalized average error is -0.03% and the normalized stan-dard deviation is 5% during the validation period. The normal-ized errors are w

30、ithin 5% range for stable pump speeds. The experiment results show that the water flow determined by the power-head-based pump flow station agrees well with the water flow meter measurement. The VFD frequency, which is related to the pump speed, was also presented in Figures 10 and 11. It was found

31、that there is a significant error when the pump speed has a sudden change. Six significant pump speed drops, from A to F, were identified in Figure 10. Each speed drops led an underesti-mated flow range (with significant negative errors) and a following overestimated flow range (with significant pos

32、itive errors). It indicates that the accuracy of the pump flow station depends on the stability of the pump speed. The stable pump speed control is mandatory to use a pump flow station. There-fore, it is recommended to commission the system to maintain stable pump speeds. More effective data can be

33、collected during the calibration and a more accurate flow can be obtained during the application. It also indicates that signifi-cant errors occur with the VFD frequency less than 35Hz, which is related to the motor power less than 7kW or 19% of nominal motor power based on Figure 5. The inaccurate

34、motor efficiency in this power range is another reason for these errors. The measured pump head and motor power always have fluctuations due to turbulent flow impact and electronic noise from VFD motor power analog output (see Figure 5). It causes the calculated flow up and down with unacceptable am

35、plitudes if instant readings are used. It is recommended to use the moving average of the measured pump head and motor power to calculated the water flow, which is used to determine the cooling energy consumption rate.After the development of the pump flow station, the cool-ing energy (rate) meter w

36、as developed by adding two chilled water temperature measurements. Figure 11 shows the measured chilled water supply temperature (CWST) and chilled water return temperature (CWRT). Figure 12 shows the measured cooling energy consumption rate as well as the water flow from the developed pump flow sta

37、tion from August 15, 2006 to August 22, 2006. The motor power and the pump head were trended with a 5-minute interval and the pump water flow is calculated based on an one-hour moving average of the motor pump and the pump head. The energy meter shows a standard HVAC system profile with a high cooli

38、ng load during the occupied time and a low cooling load during the unoccupied time. CONCLUSIONThe theory of power-head-based pump flow stations is demonstrated. The pump water flow can be determined using the motor power read from a VFD and the pump head measured by a water differential pressure sen

39、sor as well as the calibrated motor and pump efficiency. A power-head-based pump flow station was experimen-tally tested and developed on a chilled water system. The water Table 1. 50HP Motor EfficiencyRelative Power, %100 75 50 25Efficiency, %90.8 91.9 91.1 87.1Figure 8 Pump efficiency.2010, Americ

40、an Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior

41、written permission.576 ASHRAE Transactionsflow determined by the power head based pump flow station agrees well with the water flow meter measurement. The pump speed control has a significant impact on the accuracy of the pump flow station. The stable pump speed control is mandatory to use a pump fl

42、ow station. Under stable pump speeds, the measured flow errors can be controlled within 5% of the measured maximum flow (744 GPM or 2.82 m3/s) Then a cool-ing energy meter was developed using the pump flow station to monitor the cooling consumption of this chilled water system.NOMENCLATUREf=function

43、H = pump head, Psi or kPaQ = water flow, GPM or m3/sS = sum of the squares of the residualsW =power, kWa, b, c =constants= efficiencySubscriptsn = nominalREFERENCESAlger, P. L. 1970. Induction Machines. Gordon and Breach Science Publishers. ASHRAE. 2001. 2001 ASHRAE Fundamentals Handbook, Measuremen

44、t and Instruments, American Society of Figure 9 Water flow comparison.Figure 10 Normalized flow difference.2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additio

45、nal reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.2010 ASHRAE 577Heating, Refrigerating and Air-conditioning Engineers, Inc., Atlanta, GA.Fink, D. G. and H. W. Beaty. 2000. Standard Handbook for Electrical Engine

46、ers. McGraw-hill, London.Friedman, H., M. Shuman, D. Claridge, J. Curtain, and P. Haves. 2007. Building Commissioning: Innovation to Practice Technical Report. California Energy Commis-sion, PIER Energy-Related Environmental Research Program. CEC-500-2008-074.Liu M. 2002. Variable Speed Drive Volume

47、tric Tracking (VSDVT) for Airflow Control in Variable Air Volume (VAV) Systems. Journal of Solar Energy Engineering, Transactions of the ASME, v 125, n 3, p 318-323.Liu M., G. Liu, I. Joo, L. Song and G. Wang. 2005. Develop-ment of In-situ Fan Curve Measurement for VAV AHU Systems. Journal of Solar

48、Energy Engineering, Transac-tions of the ASME, v 127, n 2, May, 2005, p 287-293.McQuiston, F. C., J. D. Parker and J. D. Spitler. 2005. Heat-ing, Ventilating, and Air Conditioning Analysis and Design, 6th Ed. Chapter 10-5, System Design. John Wiley & Son, Inc.Wang, G. and M. Liu. 2005. Development o

49、f Power-head based fan airflow station, Proceedings of International Conference for Enhanced Building Operations 2005, Pittsburg, Pennsylvania.Wang, G. and M. Liu. 2007. Development of Motor Power Based Fan Airflow Station, Proceedings of Energy Sus-tainability 2007, Long Beach, California.Wildi, T. 2000. Electrical Machines, Drives, and Power Sys-tems. Prentice Hall, Upper Saddle River, NJ.Yuill