外文翻譯--離心式和往復式壓縮機的工作效率特性

1、<p><b>　　中文3906字</b></p><p>　　離心式和往復式壓縮機的工作效率特性</p><p>　　Rainer Kurz , Bernhard Winkelmann , and Saeid Mokhatab</p><p>　　往復式壓縮機和離心式壓縮機具有不同的工作特性，而且關于效率的定義也不同。本文提供了一個

2、公平的比較準則，得到了對于兩種類型機器普遍適用的效率定義。這個比較基于用戶最感興趣的要求提出的。此外，對于管道的工作環(huán)境影響和在不同負載水平的影響給出了評估。</p><p>　　乍一看，計算任何類型的壓縮效率看似是很簡單的：比較理想壓縮過程和實際壓縮過程的工作效率。難點在于正確定義適當的系統(tǒng)邊界，包括與之相關的壓縮過程的損失。除非這些邊界是恰好定義的，否則離心式和往復式壓縮機的比較就變得有缺陷了。</p&

3、gt;<p>　　我們也需要承認，效率的定義，甚至是在評估公平的情況下，仍不能完全回應操作員的主要關心問題：壓縮過程所需的驅動力量是什么？要做到這一點，就需要討論在壓縮過程中的機械損失。</p><p>　　隨著時間的推移效率趨勢也應被考慮，如非設計條件，它們是由專業(yè)的流水線規(guī)定，或者是受壓縮機的工作時間和自身退化的影響。</p><p>　　管道使用的壓縮設備涉及到往復式和

4、離心式壓縮機。離心式壓縮機用燃氣輪機或者是電動馬達來驅動。所用的燃氣輪機，總的來說，是兩軸發(fā)動機，電動馬達使用的是變速馬達或者變速齒輪箱。往復壓縮機是低速整體單位或者是可分的“高速”單位，其中低速整體單位是燃氣發(fā)動機和壓縮機在一個曲柄套管內。后者單位的運行在750-1,200rpm范圍內（1,800rpm是更小的單位）并且通常都是由電動馬達或者四沖程燃氣發(fā)動機來驅動。</p><p><b>　　效率&

5、lt;/b></p><p>　　要確定任何壓縮過程的等熵效率，就要基于測量的壓縮機吸入和排出的總焓(h)，總壓力(p)，溫度(T)和熵(s)，于是等熵效率變?yōu)椋?(Eq.1)</p><p>　　并且加上測量的穩(wěn)態(tài)質量流m，吸收軸功率為：</p><p><b>　　(Eq.2)</b></p>

6、;<p><b>　　考慮機械效率。</b></p><p>　　理論（熵）功耗（這是絕熱系統(tǒng)可能出現的最低功耗）如下：</p><p><b>　　(Eq.3)</b></p><p>　　流入和流出離心式壓縮機的流量可以視為“穩(wěn)態(tài)”。環(huán)境的熱交換通常可以忽略。系統(tǒng)邊界的效率計算通常是用吸入和排出的噴嘴。需

7、要確定的是，系統(tǒng)邊界要包含所有內部泄露途徑，尤其是從平衡活塞式或分裂墻滲漏的循環(huán)路徑。機械效率，在描述軸承和密封件的摩擦損失以及風阻損失時可以達到98%和99%。</p><p>　　對于往復式壓縮機，理論的氣體馬力也是由Eq.3給出的，鑒于吸力緩沖器上游和排力緩沖器下游的吸氣和排氣壓力脈動。往復壓縮機就其性質而言，從臨近單位需要多方面的系統(tǒng)來控制脈動和提供隔離（包括往復式和離心式），以及可以自然存在的來自管線的

8、管流量和面積管道。對于任何一個低速或高速單位的歧管系統(tǒng)設計，使用了卷相結合，管道長度和壓力降元素來創(chuàng)造脈動（聲波）濾波器。這些歧管系統(tǒng)（過濾器）引起壓力下降，因此必須在效率計算時考慮到。潛在的，從吸氣壓力扣除的額外壓力不得不包含進殘余脈動的影響。就像離心壓縮機一樣，傳熱就經常被忽視。</p><p>　　對于積分的機器，機械效率一般取為95%。對于可分機機械效率一般使用97%。這些數字似乎有些樂觀，一系列數字顯示

9、，往復式發(fā)動機機械損失在8-15%之間，往復壓縮機的在6-12%（參考1往復壓縮機招致號碼：庫爾茲，R.，K.，光布倫，2007）。</p><p><b>　　工作環(huán)境</b></p><p>　　在這樣的情況下，當壓縮機在一個系統(tǒng)中運行時，管道長度Lu上游和Ld下游，以及管道pu上游的初始壓力和管道pe下游的終止壓力均被視為常量，在管道系統(tǒng)中我們有一個壓縮機運行的

10、簡單模型（圖1）。</p><p>　　圖1：管道段的概念模型（文獻2：庫爾茲.R，M.由羅穆斯基，2006年）。</p><p>　　對于給定的，標準管線定量流動能力將在吸入階段強加壓力，在壓縮機放電區(qū)強加壓力。對于給定的管線，壓縮機站頭部（）流（Q）關系可以近似表述為 （Eq.4）</p><p>　　其中和是常數（對于一個

11、給定的管道幾何）分別描述了管道兩邊的壓力和摩擦損失（文獻2：庫爾茲.R，M.由羅穆斯基，2006年）。</p><p>　　除去其他問題，這意味著對于帶管道系統(tǒng)的壓縮機站，頭部所需流量揚程是由管道系統(tǒng)規(guī)定的（圖2）。特別地，這一特點對于壓縮機需要的能力允許頭部減量，按照規(guī)定的方式反之亦然。管道因此將不需要改變頭部的流量恒定（或壓力比）。</p><p>　　圖2：建立在4式上的機頭流量關系

12、。</p><p>　　在短暫的情況下（如包裝其間），最初的操作條件遵循恒功率分布，如頭部流量關系如下：</p><p><b>　?。‥q.5）</b></p><p>　　并將漸進地達到穩(wěn)定的關系（文獻3：奧海寧S.，R.庫爾茲，2002年）</p><p>　　在上述要求的基礎上，必須控制壓縮機輸出與系統(tǒng)要求匹配。

13、該系統(tǒng)需求的特點是系統(tǒng)流程和系統(tǒng)頭部或壓力比的強烈關系。管線壓縮機提供了在操作條件經驗下的大量變化，一個重要問題就是如何使壓縮機適應這樣變化的條件，具體的說就是如何影響效率。</p><p>　　離心壓縮機具有相當大的平頭部和流程特點。這意味著壓力比的改變對機器的實際流程有重大的影響（文獻4：庫爾茲R.，20004年）。對于一個恒速運行的壓縮機，頭部或壓力比隨著流量的增加而減少?？刂茐嚎s機內的流程可以實現壓縮機不

14、同的運行速度。這是控制離心壓縮機最便捷的方法。兩軸燃氣輪機和變速電機允許大范圍的速度變化（通常是最大速度或更多的40%或50%到100%）。應當指出，被控制的值通常不是速度，但速度是間接平衡由渦輪產生的動力（受進入燃氣輪機燃油流量控制）和壓縮機的吸收功率。</p><p>　　事實上，在過去15年安裝的任何離心壓縮機在管線服務方面是由調速器來驅使的，通常是兩軸燃氣輪機。年長的設施和服務設施在其他管線服務有時使用單

15、軸燃氣輪機（允許速度90%到100%的變化）和恒速電動機。在這些裝置中，吸節(jié)流或可變進氣導葉用來提供控制方法。</p><p>　　圖3：典型的管線運行點繪制成的典型離心壓縮機性能圖。</p><p>　　離心壓縮機的運行封套受最大允許速度限制，最小流量（涌）和最大流量（窒息或石墻）（圖3）。另一個限制因素可能是可用的驅動電源。</p><p>　　只有最小流量需要

16、特別注意，因為它被定義為壓縮機的一種氣動穩(wěn)定性的極限?？缭竭@個限制以降低流動將導致壓縮機流動逆轉，這可能會損壞壓縮機。調制解調器控制系統(tǒng)通過打開一個循環(huán)閥來控制這種情況。出于這個原因，幾乎所有的現代壓縮機裝置都使用帶有控制閥的循環(huán)線，當壓縮機內的流量趨于穩(wěn)定極限時這種控制閥允許流量的增加。控制系統(tǒng)不斷地監(jiān)測壓縮機關系喘振線的運行點，并且有必要的話自動地開關循環(huán)閥。對于大多數應用來說，帶有開放或部分開放循環(huán)閥的運行模式只被用于開啟和關閉階

17、段，或者是在混亂運行條件時的短暫時期。</p><p>　　假設由公式4得到管線特點，壓縮機的葉輪將在達到或接近其最大效率時被選出來運行，這個最大效率是由管線強加在整個系列的頭部和流量條件下的。這可能是有一個速度（N）控制的壓縮機，因為一個壓縮機的最有效點是由一種關系而連接的，這種關系需要大約（風扇法方程）：</p><p><b>　　(Eq.6)</b></

18、p><p>　　為滿足上述關系的操作點，吸入氣壓是（基于效率幾乎保持不變這個的事實）：</p><p>　　(Eq.7)　　正因為如此，這種力-速度關系允許動力渦輪運行達到或非常接近其整個范</p><p>　　圍的理想速度。管線中典型的運行方案允許壓縮機和動力渦輪在大多數時間里在最有效點運行。然而，燃氣輪機的燃氣生產商將在部分負荷運行時丟失一些熱效率。</p&g

19、t;<p>　　圖3顯示了一個典型的實際例子：不同流動要求的管線運行點繪制成用于壓縮機站中的速度控制離心壓縮機性能圖。</p><p>　　往復壓縮機將自動服從系統(tǒng)壓力比的需求，只要沒有超出機械的限制條件（桿負載功率）。系統(tǒng)吸排氣壓力的改變將僅能引起閥門或早或晚的開啟。頭部可以自動下降因為閥門可以降低排氣端的管線壓力和/或吸入端更高的管線壓力。因此，如果沒有額外的措施，流量將大致恒定——除了容積效率

20、將增加的變化，所以降低壓力比而增加流量。</p><p>　　控制的挑戰(zhàn)存在于系統(tǒng)要求的流量調整。如果沒有額外的調整，隨著壓力比的變化，壓縮機流量的改變微乎其微。從歷史上看，通過改變激活機器的數量使管線安裝許多小的壓縮機和調整流量。這個容量和負荷可通過速度調諧，或者通過一個單一單元的缸間隙中的許多小調整（加載步驟）來調諧。隨著壓縮機的發(fā)展，控制容量的負擔轉移到獨立壓縮機上。</p><p>

21、;　　負荷控制是壓縮機運行的一個關鍵組成部分。從管線操作角度來看，在機組中流量變化要符合管線投出承諾，以及實施公司最佳操作（例如，線包裝，負載預期）。從一個單元的角度來看，負荷控制包含降低單元流量（通過卸載或速度）使操作盡可能的貼近設計扭矩限制，并在壓縮機或驅動程序沒有超載的情況下進行。對于任何給定的機組入口和出口壓力，在任何負荷圖曲線上的關鍵限制都是桿負荷限制和馬力/扭矩限制。瓦斯控制通常會建立在一個機組的單元上，而這個機組運行必須達

22、到管線流量目標。地方單元控制將建立負載步驟或速度要求來限制桿負荷或達到扭矩控制。</p><p>　　改變流量的常用方法是改變速度，改變間隙，或取消激活缸頭（保持進口閥開啟）。另一種方法是卸載無限步驟，從而延緩吸氣閥封閉以減少容積效率。此外，流程的一部分可以回收或吸氣壓力可以節(jié)流從而降低質量流量，同時保持進入壓縮機的容積流量基本不間斷。</p><p>　　壓縮機控制策略應該能夠實現自動化

23、，并在壓縮機運行期間能夠簡便地調整。特別地，壓縮機設計修改的戰(zhàn)略需求（如：離心壓縮機重新旋轉，改變缸徑，或給往復壓縮機添加固定間隙）在這里不被考慮。需要指出的是，對于往復式壓縮機一個關鍵的控制要求是不超載驅動或超過機械限制。</p><p><b>　　運行</b></p><p>　　典型的穩(wěn)態(tài)管道運行將產生圖4所示的一個有效行為。該圖是評估沿管道穩(wěn)定運行特征狀態(tài)壓

24、縮機效率的結果。大中型壓縮機都將達到100%流量的最佳效率，并允許超出設計流量的10%。不同的機械效率并沒有考慮這種對比。</p><p>　　往復壓縮機效率在文獻5中被推導出，從增加的閥門效率測量與壓縮效率和造成的損失脈動衰減器。低速壓縮機的效率是可以實現的。高速往復壓縮機在效率上可能比較低。</p><p>　　圖4：以穩(wěn)態(tài)管線特性運行為基礎的在不同流量率的壓縮機效率。</p&g

25、t;<p>　　圖4顯示在較低壓力比下增加的閥門損失的影響和往復機器的較低流量，而離心壓縮機的效率幾乎保持常量。</p><p><b>　　結論</b></p><p>　　不同型號壓縮機間的效率定義和對比需要密切關注邊界條件的定義，對于這樣的邊界條件，效率和受用的運行發(fā)展趨勢同時被定義。當效率值用來計算功耗時機械效率具有重要作用。如果不考慮這些定義，

26、不同系統(tǒng)的優(yōu)缺點討論將變得不準確和有誤導性。</p><p><b>　　參考文獻：</b></p><p>　　1.庫爾茲.R.K.光布倫，2007?！巴鶑秃碗x心壓縮機的效率定義和負荷管理”</p><p>　　美國機械工程師協(xié)會 文章 GT2007-2708</p><p>　　2.庫爾茲.R，M.由羅穆斯基，200

27、6?！安粚ΨQ接壓縮機站閑置產能”。美國機</p><p>　　械工程師協(xié)會 文章 2006-90069</p><p>　　3.奧海寧.S..R.庫爾茲，2002?！皟蓹C壓縮機站的系列或平行排列”。反式。美國機械工程師協(xié)會，第124欄</p><p>　　4.庫爾茲.R，2004。“離心壓縮機性能的物理”。管道仿真利益集團。棕櫚泉，加利福尼亞</p>

28、<p>　　5.米.瓦特沙發(fā)，2003。“天然氣壓縮服務六主線壓縮機閥門的性能和耐用性試驗”。天然氣機械會議。鹽湖城，UT</p><p><b>　　原文</b></p><p>　　Efficiency And Operating Characteristics Of Centrifugal And Reciprocating Compressors&l

29、t;/p><p>　　By Rainer Kurz, Bernhard Winkelmann, and Saeid iVIokhatab</p><p>　　Reciprocating compressors and centrifugal compressors have different operating characteristics and use different eifici

30、ency definitions. This article provides guidelines for an equitable comparison, resulting in a universal efficiency definition for both types of machines. The comparison is based on the requirements in which a user is ul

31、timately interested. Further, the impact of actual pipeline operating conditions and the impact on efficiency at different load levels is evaluated.</p><p>　　At first glance, calculating the efficiency for a

32、ny type of compression seems to be straightforward: comparing the work required of an ideal compression process with the work required of an actual compression process. The difficulty is correctly defining appropriate sy

33、stem boundaries that include losses associated with the compression process. Unless these boundaries are appropriately defined, comparisons between centrifugal and reciprocating compressors become flawed.</p><

34、p>　　We also need to acknowledge that the efficiency definitions, even when evaluated equitably, still don't completely answer one of the operator's main concerns: What is the driver power required for the comp

35、ression process?To accomplish this, mechanical losses in the compression systems need to be discussed.</p><p>　　Trends in efficiency should also be considered over time, such as off-design conditions as they

36、 are imposed by typical pipeline operations, or the impact of operating hours and associated degradation on the compressors.</p><p>　　The compression equipment used for pipelines involves either reciprocatin

37、g compressors or centrifugal compressors. Centrifugal compressors are driven by gas turbines, or by electricmotors. The gas turbines used are, in general,two-shaft engines and the electric motor drives use either variabl

38、e speed motors, or variable speed gearboxes. Reciprocating compressors are either low speed integral units, which combine the gas engine and the compressor in one crank casing,or separable "high-speed" units<

39、;/p><p>　　Efficiency</p><p>　　To determine the isentropic efficiency of any compression process based on total enthalpies (h), total pressures (p), temperatures (T)and entropies (s) at suction and

40、discharge of the compressor are measured, and the isentropic efficiency r\^ then becomes:</p><p><b>　　(Eq.1)</b></p><p>　　and, with measuring the steady state mass flow m, the absorb

41、ed shaft power is:</p><p><b>　　(Eq.2)</b></p><p>　　considering the mechanical efficiency r\^.</p><p>　　The theoretical (isentropic) power consumption (which is the lowes

42、t possible power consumption for an adiabatic system) follows from:</p><p><b>　　(Eq.3)</b></p><p>　　The flow into and out of a centrifugal compressor can be considered as "stead

43、y state."Heat exchange with the environment is usually negligible. System boundaries for the efficiency calculations are usually the suction and discharge nozzles. It needs to be assured that the system boundaries e

44、nvelope all internal leakage paths,in particular recirculation paths fi^om balance piston or division wall leakages. The mechanical efficiency r)^.,, describing the friction losses in bearings and seals, as w</p>

45、<p>　　For reciprocating compressors, theoretical gas horsepower is also given by Eq. 3,given the suction and discharge pressure are upstream of the suction pulsation dampeners and downstream of the discharge pulsati

46、on dampeners. Reciprocating compressors, by their very nature, require manifold systems to control pulsations and provide isolation from neighboring units (both reciprocating and centrifugal), as well as from pipeline fl

47、ow meters and yard piping and can be extensive in nature.The design of </p><p>　　For integral machines, mechanical efficiency is generally taken as 95%. For separable machines a 97% mechanical efficiency is

48、often used. These numbers seem to be somewhat optimistic, given the fact that a number of sources state that reciprocating engines incur between 8-15% mechanical losses and reciprocating compressors between 6-12%(Ref 1:

49、Kurz , R., K. Brun, 2007).</p><p>　　Operating Conditions</p><p>　　For a situation where a compressor operates in a system with pipe of the length Lu upstream and a pipe of the length Ld downstre

50、am, and further where the pressure at the beginning of the upstream pipe pu and the end of the downstream pipe pe are known and constant, we have a simple model of a compressor station operating in a pipeline system (Fig

51、ure 1).</p><p>　　Figure 1: Conceptual model of a pipeline segment (Ref. 2: Kurz, R., M. Lubomirsky.2006).</p><p>　　For a given, constant flow capacity Qstd the pipeline will then impose a pressu

52、re ps at the suction and pd at the discharge side of the compressor. For a given pipeline, the head (Hs)-flow (Q) relationship at the compressor station can be approximated by</p><p><b>　　（Eq.4）</b&

53、gt;</p><p>　　where C3 and C4 are constants (for a given pipeline geometry) describing the pressure at either ends of the pipeline, and the friction losses, respectively(Ref 2: Kurz, R., M. Lubomirsky, 2006).

54、</p><p>　　Among other issues, this means that for a compressor station within a pipeline system, the head for a required flow is prescribed by the pipeline system (Figure 2). In particular, this characterist

55、ic requires the capability for the compressors to allow a reduction in head with reduced flow, and vice versa, in a prescribed fashion. The pipeline will therefore not require a change in flow at constant head (or pressu

56、re ratio).</p><p>　　Figure 2: Stafion Head-Flow relationship based on Eq. 4.</p><p>　　In transient situations (for example during line packing), the operating conditions follow initially a const

57、ant power distribution, i.e. the head flow relationship follows:</p><p><b>　?。‥q.5）</b></p><p>　　and will asymptotically approach the steady state relationship (Ref 3: Ohanian, S., R

58、.Kurz, 2002).</p><p>　　Based on the requirements above, the compressor output must be controlled to match the system demand. This system demand is characterized by a strong relationship between system flow a

59、nd system head or pressure ratio.Given the large variations in operating conditions experienced by pipeline compressors, an important question is how to adjust the compressor to the varying conditions, and, in particular

60、, how does this influence the efficiency.</p><p>　　Centrinagal compressors tend to have rather flat head vs. flow characteristic. This means that changes in pressure ratio have a significant effect on the ac

61、tual flow through the machine (Ref 4:Kurz, R., 2004). For a centrifugal compressor operating at a constant speed, the head or pressure ratio is reduced with increasing flow.</p><p>　　Controlling the flow thr

62、ough the compressor can be accomplished by varying the operating speed of the compressor This is the preferred method of controlling centrifugal compressors. Two shaft gas turbines and variable speed electric motors allo

63、w for speed variations over a wide range (usually from 40-50% to 100% of maximum speed or more).It should be noted, that the controlled value is usually not speed, but the speed is indirectly the result of balancing the

64、power generated by the power turbine</p><p>　　Virtually any centrifugal compressor installed in the past 15 years in pipeline service is driven by a variable speed driver, usually a two-shaft gas turbine. Ol

65、der installations and installations in other than pipeline service sometimes use single-shaft gas turbines (which allow a speed variation from about 90-100% speed) and constant speed electric motors. In these installatio

66、ns, suction throttling or variable inlet guide vanes are used to Drovide means of control.</p><p>　　Figure 3: Typical pipeline operating points plotted into a typical centrifugal compressor performance map.&

67、lt;/p><p>　　The operating envelope of a centrifugal compressor is limited by the maximum allowable speed, the minimum flow (surge flow),and the maximum flow (choke or stonewall)(Figure 3). Another limiting fact

68、or may be the available driver power.</p><p>　　Only the minimum flow requires special attention, because it is defined by an aerodynamic stability limit of the compressor Crossing this limit to lower flows w

69、ill cause a flow reversal in the compressor, which can damage the compressor. Modem control systems prevent this situation by automatically opening a recycle valve. For this reason, virtually all modern compressor instal

70、lations use a recycle line with control valve that allows the increase of the flow through the compressor if it comes ne</p><p>　　Assuming the pipeline characteristic derived in Eq. 4, the compressor impelle

71、rs will be selected to operate at or near its best efficiency for the entire range of head and flow conditions imposed by the pipeline. This is possible with a speed (N) controlled compressor, because the best efficiency

72、 points of a compressor are connected by a relationship that requires approximately (fan law equation):</p><p><b>　　(Eq.6)</b></p><p>　　For operating points that meet the above relat

73、ionship, the absorbed gas power Pg is (due to the fact that the efficiency stays approximately constant):</p><p><b>　　(Eq.7)</b></p><p>　　As it is, this power-speed relationship allo

74、ws the power turbine to operate at, or very close to its optimum speed for the entire range.The typical operating scenarios in pipelines therefore allow the compressor and the power turbine to operate at its best eflicie

75、ncy for most of the time. The gas producer of the gas turbine will, however, lose some thermal efficiency when operated in part load.</p><p>　　Figure 3 shows a typical real world example: Pipeline operating

76、points for different flow requirements are plotted into the performance map of the speed controlled centrifugal compressor used in the compressor station.</p><p>　　Reciprocating compressors will automaticall

77、y comply with the system pressure ratio demands,as long as no mechanical limits (rod load power)are exceeded. Changes in system suction or discharge pressure will simply cause the valves to open earlier or later. The hea

78、d is lowered automatically because the valves see lower pipeline pressures on the discharge side and/or higher pipeline pressures on the suction side. Therefore, without additional measures, the flow would stay roughly t

79、he same — except </p><p>　　The control challenge lies in the adjustment of the flow to the system demands. Without additional adjustments, the flow throughput of the compressor changes very little with chang

80、ed pressure ratio. Historically, pipelines installed many small compressors and adjusted flow rate by changing the number of machines activated. This capacity and load could be fine-tuned by speed or by a number of small

81、 adjustments (load steps) made in the cylinder clearance of a single unit. As compressors have grown</p><p>　　Load control is a critical component to compressor operation. From a pipeline operation perspecti

82、ve, variation in station flow is required to meet pipeline delivery commitments, as well as implement company strategies for optimal operation (i.e., line packing, load anticipation).From a unit perspective, load control

83、 involves reducing unit flow (through unloaders or speed)to operate as close as possible to the design torque limit without overloading the compressor or driver The critical limits on</p><p>　　The common met

84、hods of changing flow rate are to change speed, change clearance, or de-activate a cylinder-end (hold the suction valve open). Another method is an infinite-step unloader, which delays suction valve closure to reduce vol

85、umetric efficiency. Further, part of the flow can be recycled or the suction pressure can be throttled thus reducing the mass flow while keeping the volumetric flow into the compressor approximately constant.</p>

86、<p>　　Control strategies for compressors should allow automation, and be adjusted easily during the operation of the compressor.In particular, strategies that require design modifications to the compres.sor (for exa

87、mple: re-wheeling of a centrifugal compressor, changing cylinder bore, or adding fixed clearances for a reciprocating compressor)are not considered here. It should be noted that with reciprocating compressors, a key cont

88、rol requirement is to not overload the driver or to exceed mechanical l</p><p><b>　　Operation</b></p><p>　　The typical steady state pipeline operation will yield an efliciency behavi

89、or as outlined in Figure 4. This figure is the result of evaluating the compressor efTiciency along a pipeline steady state operating characteristic. Both compressors would be sized to achieve their best efficiency at 10

90、0% flow, while allowing for 10% flow above the design flow. Different mechanical efficiencies have not been considered for this comparison.</p><p>　　The reciprocating compressor efl'iciency is derived n-

91、om valve efficiency measurements in Ref 5 (Noall, M., W. Couch, 2003) with compression efficiency and losses due to pulsation attenuation devices added. The efficiencies are achievable with low speed compressors. High sp

92、eed reciprocating compressors may be lower in efficiency.</p><p>　　Figure 4: Compressor Efficiency af different flow rates based on operation aiong a steady state pipeline characteristic.</p><p>

93、;　　Figure 4 shows the impact of the increased valve losses at lower pressure ratio and lower flow for reciprocating machines, while the efficiency of the centrifugal compressor stays more or less constant.</p><

94、;p>　　Conclusions</p><p>　　Efficiency definitions and comparison between different types of compressors require close attention to the definition of the boundary conditions for which the efficiencies are d

95、efined as well as the operating scenario in which they are employed. The mechanical efficiency plays an important role when efficiency values are used to calculate power consumption. If these definitions are not consider

96、ed, discussions of relative merits of different systems become inaccurate and misleading. </p><p>　　REFERENCES</p><p>　　1 Kurz . R.. K. Brun. 2007. " EfTiciency Definition and Load Manageme

97、nt for Reciprocating and Centrifugal C ompressors," ASME Paper GT2OO7-27O81.</p><p>　　2 Kurz. R., M. Lubomirsky, 2006. "Asymttietric Solution for Compressor Station Spare Capacity."ASMt: Paper

98、 2006-90069.</p><p>　　3 Ohanian. S.. R. Kurz. 2002, "Series or Parallel Arrangement in a Two-Unit Compressor Station." Trans.ASME JEng for GT and Power. Vol.124.</p><p>　　4 Kurz. R.. 2

99、004. "The Physies of Centrifugal Compressor Performance." Pipeline Simulation Interest Group. Palm Springs. CA.</p><p>　　5 Noall, M.. W. Couch. 2003, "Performance and Endurance Tests of Six Ma