1、2008 ASHRAE 451ABSTRACT Two supply fan speed control methods for a variableairflow volume (VAV) system using the fan airflow station(FAS) are introduced in this paper. In the first method, thesupply fan speed is controlled to maintain an optimized systemresistance (S), which is calculated based on t
2、he measured fanhead, and airflow. In the second method, the supply fan speedis controlled to maintain the duct static pressure set point,which is reset based on the airflow ratio measured by the FAS.Both methods can make the terminal box open more comparedwith the conventional supply fan speed contr
3、ol using a fixedduct static pressure set point. These innovative methods can beapplied to systems with either direct digital control (DDC) orpneumatic control terminal boxes. The case studies show thatthe new control methods can significantly save supply fanpower and improve fan efficiency.INTRODUCT
4、IONThe ASHRAE handbook 1999 introduces severalcontrol methods for the supply fan in VAV systems. Flow maybe modulated in a VAV system by using dampers on the outletside of the fan, inlet vanes on the fan, vane-axial fans withcontrollable pitch fan blades, or variable-speed control of thefan motor. T
5、ypically, the inputs to any of these controlleddevices are modulated in order to maintain a duct static pres-sure set point. The pressure sensor must be properly placed tomaintain optimum pressure throughout the supply duct. Expe-rience indicates that performance is satisfactory when thesensor is lo
6、cated at 75% to 100% of the distance from the firstto the most remote terminal ASHRAE Application Hand-book, 1999. The pressure selected provides a minimum staticpressure to all air terminal units during all supply fan designconditions. Multiple static sensors are required when morethan one duct run
7、s from the supply fan. The sensor with thehighest static requirement controls the fan. Because duct run-outs may vary, a control that uses individual set points for eachmeasurement is preferred.In a single-duct VAV system, the duct static pressure setpoint is typically selected by the designer to be
8、 a fixed value.The sensor should be located in the ductwork where the estab-lished set point ensures proper operation of the zone VAVboxes under varying load (supply airflow) conditions. A short-coming of this approach is that static pressure control is basedon the readings of a single sensor that i
9、s assumed to representthe pressure available to all VAV boxes. If the design or appli-cation of the sensor malfunctions, or the sensor is placed in alocation that is not representative, operating problems willresult. Another issue is the static pressure set point, which isnormally set at a very high
10、 value to meet the design conditions.However, the building is operated under off-peak conditionsmost of the time. For a fixed static pressure set point, all of theVAV boxes tend to close as the zone loads and flow require-ments decrease. Therefore, the flow resistance increases withdecreasing loads
11、and consumes a significant amount of fanpower.An alternative approach to supply fan control in a VAVsystem uses flow readings from the direct digital control(DDC) zone terminal boxes to integrate zone VAV require-ments with supply fan operations. Significant fan energysavings are possible if the sta
12、tic pressure set point is reset sothat at least one of the VAV boxes remains open. With thisapproach, the duct system resistance remains similar range.Englander and Norford 1992, Hartman 1993, and Warrenand Norford 1993 proposed several different strategies basedSupply Fan Control Methods for VAV Sy
13、stems Using a Fan Airflow StationGuopeng Liu, PhD Mingsheng Liu, PhD, PEAssociate Member ASHRAE Member ASHRAEGuopeng Liu is a senior project engineer of Building Energy Solutions and Technology, Bes-Tech, Inc., Dallas, TX. Mingsheng Liu is aprofessor in the Department of Architectural Engineering, U
14、niversity of Nebraska Lincoln, Omaha, NE.NY-08-0552008, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions, Volume 114, Part 1. For personal use only. Additional reproduction, distribution, or transmission in either prin
15、t or digital form is not permitted without ASHRAEs prior written permission.452 ASHRAE Transactionson this concept. Englander and Norford used simulations toshow that either static pressure or fan speed can be controlleddirectly using a flow error signal from one or more zones andsimple rules. Their
16、 technique forms the basis of the reset strat-egy given below. At each decision interval (e.g., 5 minutes), the followinglogic can be applied:1. Check the controller outputs for representative VAVboxes and determine time-averaged values over the lastdecision interval.2. If any of the controller outp
17、uts are greater than a thresh-old value (e.g., 98%), then increase the static pressure setpoint by a fixed value (e.g., 5% of the design range) andgo to Step 4. Otherwise, go to Step 3.3. If all of the controller outputs are less than a thresholdvalue (e.g., 90%), then decrease the static pressure s
18、etpoint by a fixed value (e.g., 5% of the design range) andgo to Step 4. Otherwise, do not change the set point.4. Limit the set point between upper and lower limits basedon upper and lower flow limits and the duct design.Unfortunately, this method cannot be used in buildingsthat have a pneumatic te
19、rminal box controller. Box malfunc-tion, communication loss, and other issues make this methoddifficult to implement, even for DDC systems. How should system resistance be maintained relativelyconstant for the pneumatic terminal box controller? Thesystem resistance can be calculated as the ratio of
20、fan head tothe square of the airflow. Accurate measurement of airflow inall ranges is necessary and presents a major challenge incontrolling the system resistance at an optimized value.The most common method is to measure the airflowsusing flow stations in the main supply duct and in the mainreturn
21、duct. For accuracy within 5% to 10%, the airflowmeasurement station requires a straight duct for 6 to 10 ductdiameters upstream and 3 duct diameters downstream NEBB1986. Unfortunately, there are very few systems that havesuch duct runs in the main supply and return ducts. Moreover,the air dynamic he
22、ad varies proportionally to the square of theair velocity. When the velocity is reduced to a lower range, thepressure transducer often cannot provide adequate accuracyfor the dynamic head measurement.To increase the accuracy, it is recommended to measurethe airflow at the location with the highest v
23、elocity Kettler1995. The fan inlet is such a place. Fan inlet technology canmeasure the airflow at the fan inlet mounted in the intake bellof the fan. However, the airflow profile in the fan inlet varieswith the total airflow, which results in high uncertainty anddifficulty of measurement. This meth
24、od cannot provide theaccuracy required for the volumetric tracking. The cost of theturbo meter is another limitation.The thermal anemometer or hot-wire anemometer hasbeen used to determine the air velocity at a point in the flowfield. The measurement range is from 0.005 m/s (0.98 ft/min)to 5 m/s (98
25、4 ft/min), and the accuracy is 2% to 5% of the read-ing over the entire velocity ASHRAE 2001. Airflow can bemeasured accurately when an adequate number of sensors areused across a duct cross-section. However, these sensors arevery sensitive to the surface property. A sensor with a dirtysurface may s
26、end more discrepant signals than a sensor witha clean surface under the same air velocity. Installation of anadequate numbers of sensors can also be expensive. Pitot-static tubes are often used in airflow measurementsin the terminal box. The EMCS then sums all airflow measure-ments for the terminal
27、boxes. Ideally, airflow should rangefrom 0.9 m/s (177 ft/min) to 50 m/s (9842 ft/min) with an accu-racy of 1% to 5%. The main drawback is that the accuracy fallsoff at the low end of the range, where the terminal box oftentends to operate. An airflow control named VSD volumetric tracking(VSDVT) has
28、been recently developed by Liu 2002. The fanairflow station measures fan airflow using fan head, fan speed,and fan curve. VSDVT was initially used for building pressurecontrol. The fan airflow station uses the fan speed and fan headas inputs. It determines the airflow using the in-situ fan curvebase
29、d on measured fan speed, fan head, and the pre-deter-mined relationship of fan head and fan airflow (fan curve)under a given fan speed. Please refer to the Appendix fordetails.METHODOLOGYTwo control methods were developed for the supply fanspeed control using the FAS. These two methods can beutilize
30、d by systems with a DDC or pneumatic terminal boxcontroller. Method I: Control the Supply Fan Speed to Maintain a Constant System Resistance Its hard to find the optimized value for the system resis-tance set point. In practice, we can make the critical zoneterminal box fully open by adjusting the t
31、hermostats for thosezones to full cooling mode. The total airflow rate and fan headcan be measured and the system resistance can be attainedusing Equation (1). S=H/Q2(1)whereH = measured fan head, in. H2O or PaQ=airflow rate which can be measured by FASThe desirable system resistance set point shoul
32、d be a littlehigher than the measured system resistance of the actual oper-ation point when critical zone terminal box are fully open. Thiswill ensure that an adequate amount of airflow can be suppliedto each zone. We can also consider this method as a differentialpressure (DP) reset- controlling su
33、pply fan head proportion-ally to the square of airflow rate.ASHRAE Transactions 453Method II: The Fan Speed Is Controlled to Maintain the Duct Static Pressure Set Point. The Static Pres-sure Set Point Is Reset Based on the Supply Airflow Ratio, Which Is Measured by the FASThe supply air fan is contr
34、olled to maintain the duct staticpressure at the set point. When the cooling load is low, the supplyairflow rate is reduced. Consequently, the pressure loss throughthe duct will be reduced. Duct static pressure will be reset at alower value according to the airflow ratio measurements.For constant st
35、atic pressure control, extra fan head isconsumed to maintain the constant static pressure set point whenterminal box dampers downstream of the sensor are closed toconsume the fan head. The duct static pressure reset can beapplied if the pressure sensor is located at the upstream or themiddle of the
36、main duct rather than at the end location. The resetlowers the duct static pressure, allowing the terminal box damp-ers to open more fully. Therefore, the amount of supply airflowto the space for the reset will be the same as for the constant staticpressure control during the same load conditions.Th
37、e static pressure required through ductwork and boxesis proportional to the square of the supply airflow ratio. (2)whereP = a calculated static pressure set point in. H2O or PaPmax= a measured static pressure when all the boxes are open and fan is running at full speedPref= the reference pressure wh
38、ich can be 15% to 30% of the design static pressure depending on the load distribution profileQ = the calculated supply airflow rate, cfm (L/s)Qd= the measured airflow rate at design conditionPmin= the minimum static pressure set point, which is often determined based on the terminal box propertyIn
39、reality, it is difficult to obtain measured data under thedesign conditions. Therefore, Pmax and Qdin Equation 2 can beobtained by measuring the actual static pressure and supplyairflow when most terminal boxes are fully open and fan isrunning at full speed. An additional pressure Prefis added intot
40、he calculation due to different load profiles in different zones.If all the zones have the similar load ratios, 15% is recom-mended. If the zone load ratios are very diverse, a higher valueis recommended. This will ensure that zones that have highercooling loads can be supplied with enough airflow.
41、The mini-mum static pressure set point can be different and should beadjustable based on the system characteristics. For a typicalsystem, 0.3 to 0.5 in w.g. (75 Pa to 125 Pa) is recommended.Figure 1 compares the working point of the fan controlusing a fixed static pressure set point, and dynamic res
42、et staticpressure set point. With constant static pressure set point, theterminal boxes are force to close more at partial load conditionwith higher fan head. Compared with a fixed static pressure setpoint, the dynamic reset schedule will make the terminal boxesopen more with a lower fan head and sy
43、stem resistance. Thefan working point moves down with the same airflow rate,which has lower speed as shown in Figure 1. Both methods actually are pressure reset control algo-rithm using FAS. The difference is method I uses the differ-ential pressure cross the supply fan inlet and outlet (fan head),a
44、nd method II uses the static pressure sensor in the down-stream of the ductwork. For the system with more than oneduct runs, method I is recommended. CASE STUDYCase Study One (Using Method I)The case study building, located in Omaha, Nebraska,was originally built in 1960. This office building includ
45、es a 16-story tower and a two-story addition to the east. The totalconditioned area is 216,000 ft2(20,066 m2). On average, thebuilding is occupied by 400 people approximately 50 hoursper week.The Cloud Room, which is served by AHU 16 with a condi-tioned area of 5000 ft2 (465m2), has a total of 12 te
46、rminal boxesand four thermostats on the 16thfloor. The Cloud Room is usedfor meetings or parties. It has a dual-duct system with a constantvolume pneumatic-control box. Englander and Norfordsmethod 1992 cannot be implemented for this system.The cold deck and hot deck damper at the terminal box arein
47、terlocked using one actuator. The supply fan is 15 HP(11 kW) and the return fan is 5 HP (3.7 kW) without variablefrequency drive (VFD) installed. The building pressure setpoint is 0.1 in. w.g. (25 Pa), which is controlled by relief airdamper. The design airflow is 9,760 CFM (4,606 L/s) and thedesign
48、 fan head is 4.6 in. w.g. (1,146 Pa).P MAX PmaxPref()QQd-2PrefPmin,+=Figure 1 Comparison of the fan operation points using aconstant static pressure setpoint and usingmethod II.454 ASHRAE TransactionsThe system retrofit includes adding a control damper onthe main hot deck and installing a VFD on bot
49、h the supply andreturn fans. A fan head-based FAS was installed to measure thesupply and return airflow rate. The schematic diagram of AHU16s existing and added sensors is shown in Figure 2.After the continuous commissioning (CC) implementa-tion, the supply fan speed was controlled following the rulesbelow.Summer Mode. When the outside air temperature (OAT)is higher than 60oF (15.6C), the system will be switched to asingle-duct system by totally shutting off the hot deck damper.Then the supply fan speed can be controlled to maintain theconstant syst
