畢 業(yè) 設 計 (論 文 ) 英 文 翻 譯院（系）： 機電工程學院專 業(yè)： 機械設計制造及其自動化學生姓名：學 號： 指導教師： 第 1 頁混合動力液壓挖掘機動力系統(tǒng)控制戰(zhàn)略肖清 1 王慶豐 1 張彥廷 21、浙江大學流體傳動及控制國家重點實驗室，杭州，310027，中國2、中國石油大學機械電氣工程學院，東營, 257061，中國2007 年 5 月 21 日概述曾成功應用于汽車行業(yè)的混合動力系統(tǒng)，現(xiàn)正引入到液壓挖掘機中。本課題的主要重點是研究液壓挖掘機混合動力系統(tǒng)的控制策略。首先是對混合動力液壓挖掘機的結(jié)構和工作條件進行了分析。在分析的基礎上，名為發(fā)動機固定工作點控制策略被提出并在模擬實驗系統(tǒng)中研究。名為雙工作點的控制策略的提出克服了恒定工作點控制策略的局限性。雙工作點控制策略的特點和實驗結(jié)果表明，發(fā)動機的效率和電容器的充電狀態(tài)（SOC）的不能同時優(yōu)化。因此，動態(tài)調(diào)節(jié)發(fā)動機的工作點的控制策略能使系統(tǒng)更好地工作。實驗結(jié)果表明，動態(tài)工作點控制策略，可以提高發(fā)動機的工作點分布，抑制電容的SOC但是對系統(tǒng)的性能影響不大。關鍵詞：混合動力系統(tǒng);挖掘機;發(fā)動機固定工作點控制策略;雙工作點控制策略; 動態(tài)工作點控制策略1、前言能源消耗和污染在全球范圍內(nèi)越來越嚴重。由于工程機械高能耗和不良排氣，因此，對其的節(jié)能研究是十分必要和緊迫的，特別是液壓挖掘機。如果沒有重大的技術突破，那么液壓挖掘機的傳統(tǒng)能源的節(jié)能方法就不能大規(guī)模的提第 2 頁高影響力 [1,2]。從不同的工作條件下的液壓挖掘機（狀態(tài)數(shù)據(jù)從實際工作中派生）可以得出結(jié)論，其負載功率在較大范圍內(nèi)發(fā)生周期性的變化，從而發(fā)動機的工作狀態(tài)也周期性變化，因此發(fā)動機不能始終保持在一個高效率的狀態(tài)。這是液壓挖掘機的燃油經(jīng)濟性低的主要原因?；旌蟿恿ο到y(tǒng)，該系統(tǒng)包括一個發(fā)動機和一個電動馬達，發(fā)動機工作在最佳效率范圍來提高燃油經(jīng)濟性的潛力已成功地應用于車輛。因此，為達到節(jié)約能源，液壓挖掘機配備混合動力系統(tǒng)成為了一種新的方式。最近，液壓挖掘機的結(jié)構，控制策略和混合動力系統(tǒng)的能源管理的研究已經(jīng)開展 [3-9]。其中，控制策略，它直接決定在動力系統(tǒng)中的元件的工作狀態(tài)并最終影響液壓挖掘機的能源消耗，這是關心的主要問題之一。本文主要涉及液壓挖掘機混合動力系統(tǒng)的控制策略。我們逐步提出這些控制策略。當混合動力系統(tǒng)的實現(xiàn)時，負載功率波動被蓄能動力系統(tǒng)吸收，使發(fā)動機輸出平均負載功率。因此，發(fā)動機在一個恒定的高效率點工作的控制策略，可實現(xiàn)增加發(fā)動機和系統(tǒng)的效率與效益。然而，根據(jù)控制策略在恒定的工作效率高的觀點，發(fā)動機選擇工作的能力不能和平均負載功率完全相同，經(jīng)過一個工作周期負載蓄能器（SOC）的狀態(tài)會上升或下降。經(jīng)過長時間的工作，SOC將超出其工作范圍，系統(tǒng)將不再正常工作。為了克服這種局限性，我們可以采用雙工作點控制策略，即發(fā)動機工作在一個高功率點和一個高效率區(qū)域的低功率點。蓄能器的SOC超過指定的上限時，發(fā)動機切換到低功率點;當蓄能器的SOC超過指定的下限時，發(fā)動機切換到它的高功率點。發(fā)動機的效率以這種方式保持穩(wěn)定較高，蓄能器的SOC將不超過其工作范圍。在雙工作點控制策略下，如果累加器的分配工作范圍很窄，考慮系統(tǒng)的穩(wěn)第 3 頁定性，發(fā)動機頻繁的在這兩個工作點之間切換，這是不可取的。另一方面，如果累加器的SOC的工作范圍設置廣泛，累加器的效率和循環(huán)壽命將降低。因此，在我們的實驗室，動態(tài)調(diào)節(jié)發(fā)動機的工作點這個控制策略已發(fā)展到可以克服這個缺點。這種控制策略下，根據(jù)累加器的SOC發(fā)動機的工作點在高效率范圍內(nèi)動態(tài)變化，也可避免在雙工作點控制策略中遇到的問題。本文組織如下。第2節(jié)致力于混合動力系統(tǒng)的結(jié)構和工作條件。第3節(jié)發(fā)動機恒定工作點的控制策略。第4節(jié)發(fā)動機雙工作點的控制策略。第5條與實驗結(jié)果一起討論發(fā)動機動態(tài)工作點的控制策略。最后，結(jié)論是在第6節(jié)。2、動力系統(tǒng)的結(jié)構和工作條件2.1、動力系統(tǒng)的結(jié)構圖 1 并聯(lián)式混合動力液壓挖掘機的示意圖動力系統(tǒng)的結(jié)構如圖1，發(fā)動機和電動馬達用并聯(lián)混合方式來驅(qū)動液壓泵。與串行混合動力系統(tǒng)發(fā)動機機械動力直接驅(qū)動液壓泵相比能源轉(zhuǎn)換損失降低。電動機，它既有電動機的功能也可以作為發(fā)電機工作，輸出能量連同引擎或?qū)l(fā)動機多余的機械能轉(zhuǎn)換成電能，并存儲在電容器。第 4 頁2.2、動力系統(tǒng)的工作條件圖 2、挖掘工作條件下電源系統(tǒng)的輸出功率圖2顯示了動力系統(tǒng)的歸一化輸出功率（P/ P max） ，其中 P是液壓挖掘機的輸出功率， Pmax是發(fā)動機的額定功率。數(shù)據(jù)來自某些液壓挖掘機挖掘的實際工作周期。從圖中可以看出，輸出功率波動較大并具有周期性，周期時間大約只有18秒。因此，具有快速充放電速度和周期壽命長的電容，在動力系統(tǒng)中被用作蓄能器來快速平衡功率波動。3、發(fā)動機固定工作點控制策略3.1、控制策略的詳細信息根據(jù)上述分析，液壓挖掘機的負載功率是周期和循環(huán)的。在一個周期內(nèi)的負載能力，可以采取兩個組成部分：平均值和波動。因此，它是混合動力液壓挖掘機合理的利用發(fā)動機固定工作點（恒轉(zhuǎn)速和恒轉(zhuǎn)矩）的控制策略，發(fā)動機在一個固定點工作輸出的平均負載功率，波動功率由電動機電容器提供電源。這樣發(fā)動機可以具有較高的燃油經(jīng)濟性和低排放的性能始終工作在高效率范圍第 5 頁內(nèi)。在控制策略的控制下，發(fā)動機轉(zhuǎn)速恒定工作點是一個預設值。由于電動機與發(fā)動機同軸連接，其轉(zhuǎn)速和發(fā)動機相同。從圖1中可以看出。發(fā)動機的扭矩是液壓泵和電動機的轉(zhuǎn)矩的差。當負載變化時，我們應調(diào)整電動機的轉(zhuǎn)矩，保持發(fā)動機的扭矩恒定。這可以通過改變通過調(diào)節(jié)轉(zhuǎn)速同步電動機的轉(zhuǎn)差率實現(xiàn)。圖 3、電動機的機械特性曲線圖3顯示了電動機的機械特性曲線。圖中的符號有以下幾種:n 轉(zhuǎn)速M 轉(zhuǎn)矩nm 電動機的實際轉(zhuǎn)速轉(zhuǎn)差率?在這里，n m是一個常數(shù)。正如圖中所示，電動馬達的同步轉(zhuǎn)速變化時，機械特性曲線向上或向下移動和電動馬達的輸出扭矩的變化是相對應的。當同步轉(zhuǎn)速比n m低， 變?yōu)樨?，電動馬達的扭矩也變?yōu)樨摂?shù)（電動馬達作為發(fā)電機） 。第 6 頁否則 和電動馬達的扭矩是正數(shù)。 和電動機的轉(zhuǎn)矩之間的關系是由電機的n?n?機械特性曲線決定的。圖 4、控制發(fā)動機扭的矩控制框圖圖4是發(fā)動機在恒定的工作點的扭矩控制策略框圖。圖中使用的符號如下：Mei 發(fā)動機的額定扭矩nmi 電動機的額定同步轉(zhuǎn)速nmo 電動機的實際同步轉(zhuǎn)速電動機的轉(zhuǎn)差率m?Mm 電動機的輸出轉(zhuǎn)矩Meo 發(fā)動機的輸出扭矩 通過控制算法（PID選擇）給出發(fā)動機的額定扭矩M ei，電動機的額定同步轉(zhuǎn)速; 同步電動機速度控制是由一個矢量控制器控制； nmo和n m之間的差異 和mn?電動機輸出扭矩M m由n m決定；然后發(fā)動機輸出扭矩M eo來配合電動馬達驅(qū)動液壓泵。第 7 頁3.2、實驗系統(tǒng)圖 5、實驗系統(tǒng)的示意圖圖5所示是建立在我們的實驗室的一個模擬實驗臺，研究的是混合動力系統(tǒng)的控制策略。比例溢流閥是用來模擬混合動力系統(tǒng)的負載壓力。交替液壓泵的排量實現(xiàn)了負載流量的模擬。圖中的符號有以下幾種：pp 液壓泵的壓力 Q 液壓泵的流量 M1 Mot1的轉(zhuǎn)矩M2 Mot2的轉(zhuǎn)矩 U 電容器的電壓 I 電容器的電流f1 Inv1控制信號的頻率f2 Inv2控制信號的頻率qc 液壓泵排量控制信號pc 比例溢流閥壓力控制信號第 8 頁為方便控制，我們使用一臺37kW的變頻電機MOT1，它是變頻器INV1控制由圖1中發(fā)動機代替。Mot2是功率為22kW的可變頻率電動機，由變頻器INV2控制。并行連接MOT1和Mot2驅(qū)動液壓泵。一個電容為12.5 F、最大電壓400伏的電容器組，被用來作為實驗系統(tǒng)的蓄能器。該系統(tǒng)的主要控制單元由工業(yè)控制計算機，數(shù)據(jù)采集卡和一個數(shù)據(jù)控制卡組成合適的傳感器，用于測量PP， n， Q，M 1，M 2，U，I等?？刂破魇占吞幚韥碜詡鞲衅鞯臄?shù)據(jù)，并輸出控制信號f 1， f2，q c，p c控制轉(zhuǎn)速電動馬達和液壓系統(tǒng)的壓力流量。3.3、控制策略的實驗結(jié)果圖 6、液壓泵的壓力和流量根據(jù)分析，用上文所述的實驗系統(tǒng)對發(fā)動機在固定工作點控制策略進行研究。圖6顯示了在一個工作周期的液壓泵的流動速率(Q/ Q max)和壓力（p/ p max）（數(shù)據(jù)來自實際工作循環(huán)的液壓挖掘機） 。我們將流速和壓力轉(zhuǎn)換為相應q c和p c來控制實驗中液壓泵和比例溢流閥。第 9 頁圖 7、輸出功率的比較圖7給出了標準化Mot1、 Mot2和電容的輸出功率（P/ P max）的比較?？梢钥闯觯诠ぷ髦芷谥蠱OT1輸出功率波動小，說明發(fā)動機工作點幾乎是恒定的，而Mot2 的輸出功率是波動的。圖7還顯示，輸出功率Mot2總是低于電容器，它們之間的區(qū)別是電源轉(zhuǎn)換損失。圖7顯示發(fā)動機恒定的工作點的控制策略基本上是可行的，但MOT1 的輸出功率不是完全不變。其原因是發(fā)動機扭矩控制算法是一種簡單的PID，是不夠準確的。提高控制算法是我們下一步研究的重點。4、發(fā)動機雙工作點控制策略由于選擇的發(fā)動機工作電源能力和平均負載功率不完全一樣，SOC的電容器長時間工作將超過其工作范圍。我們進一步制定了一種控制策略，當SOC超過其上限的時候，發(fā)動機切換到一個在高效率范圍內(nèi)的低功率的工作點，當SOC到其下限，發(fā)動機切換到一個在高效率范圍內(nèi)高功率的工作點，并命名為發(fā)動機雙工作點控制策略。在我們上面的實驗系統(tǒng)提到的雙工作點控制策略的研究。其控制方法是恒定的工作點控制策略，即電動機的扭矩恒定是通過調(diào)節(jié)電動機的同步轉(zhuǎn)速。實第 10 頁驗曲線如圖8，大功率發(fā)動機的工作點是P h，低功耗的工作點是P l ，P/P max是MOT1額定輸出功率，S是電容的SOC 。這個數(shù)字說明根據(jù)電容的 SOC，Mot1工作點在P l和P h之間切換，交換機的特點與上述分析是一致的，這表明這種控制策略的可行性。在發(fā)動機恒定的工作點的控制策略中，該控制策略不能穩(wěn)定在P l和P h工作點之間的一個恒定值上。圖 8、雙工作點控制策略的實驗曲線由此可以推斷的實驗結(jié)果，如果電容的SOC的工作范圍窄，該發(fā)動機將在兩個工作點之間切換頻繁，這是不利于系統(tǒng)的穩(wěn)定工作的。如果電容的SOC的工作范圍很廣，電容器的工作效率和工作環(huán)境將降低。因此，動態(tài)調(diào)整發(fā)動機的工作點的控制策略，被用于優(yōu)化發(fā)動機的工作狀態(tài)和電容的SOC。5、發(fā)動機動態(tài)工作點控制策略5.1、 控制策略的詳細內(nèi)容在此控制策略下，根據(jù)電容器的每一個工作周期后的SOC來動態(tài)調(diào)整發(fā)動機的工作點。有兩種控制策略的目標：一個是確保發(fā)動機的工作點分布在其高效率的范圍內(nèi)或附近。另一種是抑制變幅電容的SOC的變化范圍?？刂撇呗匀缦滤?。第 11 頁圖 9、發(fā)動機的效率圖第1步：預估負載所需的平均功率，確定發(fā)動機的高功率和低功率限，高、低功率限所確定的區(qū)域與發(fā)動機高校區(qū)的重疊部分為其工作區(qū)，即圖9所示虛線所覆蓋的H區(qū)。圖9中的坐標標為轉(zhuǎn)速（n /n max ）和轉(zhuǎn)矩(M /Mmax).。第2步：根據(jù)負載所需平均功率在H區(qū)選擇發(fā)動機的初始工作點P0（n e，M e） 。第3步：設置電容初始SOC值 S0及其靈敏度 tS?第4步： i（i = 1，2，3 .）工作周期后，如果SOC當前值 Si和SOC的前一個周期的值 滿足式（1）和（2） ，則系統(tǒng)不改變數(shù)值繼續(xù)工作，否則，調(diào)整?iS發(fā)動機的工作點，其方法如式（3） 。, (1)tiiSS???1, (2)ti0, (3)????????????? 011 &;,, SSKMnPn iiideiei其中：Pi+1(ne, Me) 發(fā)動機的工作點后，第i個工作周期Pi(ne, Me) 發(fā)動機的工作點后，第 i-1個工作周期 第 12 頁Kc 發(fā)動機功率過高時的調(diào)整系數(shù) Kd 發(fā)動機功率過低時的調(diào)整系數(shù)SOC的變化值, 等于S i?Si ? 1 S?圖 10、控制策略的流程圖第 13 頁第5步：如果有必要，發(fā)動機的工作點沿等功率線移動到H 區(qū)或附近（圖9所示） 。第6步：根據(jù)發(fā)動機的工作點的改變來調(diào)節(jié)液壓系統(tǒng)的控制信號，來滿足負載要求。第7步：當發(fā)動機的工作點沿等功率線調(diào)節(jié)，若液壓系統(tǒng)的控制信號做相應的調(diào)整后不再控制范圍內(nèi)，應犧牲發(fā)動機的效率 來滿足負載的需要?？刂撇呗缘牧鞒虉D如圖10. 在發(fā)動機等功率線上調(diào)整其轉(zhuǎn)速N e和轉(zhuǎn)矩M e將工作點P i+1（ ne，M e）調(diào)至 ，應滿足下面列出的條件：??einP,1??, (4)eM??, (5)1?iiQ其中：發(fā)動機的工作點調(diào)整后的轉(zhuǎn)速和轉(zhuǎn)矩en?,發(fā)動機工作點調(diào)整前液壓泵輸出流量1?iQ發(fā)動機工作點調(diào)整后液壓泵輸出流量?i和 , (6)11??ieiqnQ, (7)??ii其中：:qi+1 發(fā)動機工作點調(diào)整前液壓泵的排量發(fā)動機工作點調(diào)整后液壓泵的排量1??i由式(4)-(7),當發(fā)動機的轉(zhuǎn)速調(diào)節(jié)為 時，發(fā)動機的控制力矩 應是en? eM?, (8)eM??而液壓泵的排量需調(diào)整為:第 14 頁, (9)11????ieiqn液壓泵的排量可通過調(diào)節(jié)揉機制來控制?？梢酝ㄟ^調(diào)節(jié)速度調(diào)節(jié)裝置控制發(fā)動機的轉(zhuǎn)速,發(fā)動機的扭矩是:, (10)mpeM??從式（10）中可以看出。可以通過改變電動機的輸出轉(zhuǎn)矩 調(diào)節(jié)發(fā)動機m的扭矩 。也提到了可以通過調(diào)整同步電動機的轉(zhuǎn)速來實現(xiàn)發(fā)動機恒定的工eM作點控制策略中 的變化。 m因此,控制策略可通過以控制發(fā)動機的轉(zhuǎn)速,同步電動機轉(zhuǎn)速和揉機制實現(xiàn)液壓泵的流量。5.2、控制策略的實驗結(jié)果5.2.1、發(fā)動機的工作點分布圖 11、無混合動力系統(tǒng)的發(fā)動機工作分布點第 15 頁圖 12、混合動力系統(tǒng)的發(fā)動機工作分布點圖11顯示僅發(fā)動機驅(qū)動液壓系統(tǒng)時的發(fā)動機工作點分布。隨著負載的波動，發(fā)動機的工作點也伴隨著各種效率改變而改變。因此，該系統(tǒng)的效率不可能很高。圖12說明了混合動力系統(tǒng)驅(qū)動液壓系統(tǒng)系統(tǒng)時的發(fā)動機工作點分布。與圖11上所顯示的不同，圖12中發(fā)動機工作重點集中在高效率工作點分布區(qū)與所需的控制策略是一致的。圖11和12中的坐標都和圖9相同。5.2.2、電容的 SOC 的變化圖 13、電容的 SOC 變化曲線第 16 頁圖13電容的SOC在5個工作周期的變化曲線。由此可以看出，通過對發(fā)動機工作點的動態(tài)調(diào)整，電容的SOC雖有變化，但在一個小范圍內(nèi)，SOC幾個周期后趨于穩(wěn)定。由于液壓挖掘機的工作是周期性的，圖13可以推導出，SOC會穩(wěn)定在某一個值使得的電容器和系統(tǒng)工作很長時間。5.2.3、響應性能圖 14、流量響應性能的比較當發(fā)動機完全驅(qū)動系統(tǒng)時，可以通過控制液壓泵的排量和比例溢流閥的壓力實現(xiàn)模擬負載。當系統(tǒng)在混合方案的控制策略帶動下，應同時控制發(fā)動機的轉(zhuǎn)速和電動機的轉(zhuǎn)速。圖14顯示了在兩個不同的驅(qū)動方法，其中 為液maxQ壓泵的歸一化的流率。雖然混合驅(qū)動更復雜并且需要更多的控制變量，但是在這兩種動態(tài)控制策略驅(qū)動方式下的流量響應變化小.6、結(jié)論在本文中，對發(fā)動機固定工作點控制策略進行了分析。實驗結(jié)果表明，發(fā)動機固定工作點控制策略基本上可以保持發(fā)動機工作在電動機調(diào)整的恒功率下，但不能保證以后很長一段時間電容的SOC工作在所需的工作范圍內(nèi)。然后雙工第 17 頁作點控制策略克服了恒定工作點控制策略的不足之處。實驗結(jié)果表明，該控制策略可以保持SOC在以后很長一段時間工作在一個理想的工作范圍內(nèi)，但它不能使系統(tǒng)穩(wěn)定和電容在處高效率狀態(tài)下，并會造成電容快速充放電的不良現(xiàn)象。最后，以消除雙工作點控制策略弊端為目的，動態(tài)調(diào)整發(fā)動機的工作點的控制策略被提出并實驗研究。這種控制策略下，發(fā)動機的工作點是保持在高效率的區(qū)域內(nèi)或附近，電容的SOC限制在一個狹小的區(qū)域，從而避免了快速充放電并且提高電容器的使用壽命。盡管該系統(tǒng)變得更加復雜并需要更多的控制變量，但流量響應變化不大。實驗結(jié)果表明，這種控制策略的可行性。這個研究提供了一種新的混合動力系統(tǒng)和可行的控制策略。這個研究有改善目前液壓挖掘機的效率的潛力。參考文獻[1]. Y.T. Zhang, Q.F. Wang, Q. Xiao, Simulation research on energy saving of hydraulic system in hybrid construction machinery, Proceedings of the Sixth International Conference on Fluid Power Transmission and Control, 2005, pp. 509–513, ISBN:7-5062-7402-7, Hangzhou.[2]. Q.F. Wang, Y.T. Zhang, Q. Xiao, Evaluation for energy saving effect and simulation research on energy saving of hydraulic system in hybrid construction machinery, Chinese Journal of Mechanical Engineering 41 (12) (2005) 135–140.[3]. M. Matsubara, N. Matoba, Hybrid construction machine, JP Patent No.2004011256, 2004-01-01.[4]. M. Naruse, N. Oji, Hybrid type construction machine, JP Patent No.2004011502, 2004-01-15.[5]. S. Riyuu, M. Tamura, M. Ochiai, et al., Hybrid construction machine, JP Patent No.2003328397, 2003-11-19.[6]. M. Naruse, M. Tamaru, K. Kimoto, Hybrid construction equipment, US Patent No.6708787, 2004-03-23.[7]. T. Nanjo, E. Imanishi, M. Kagoshima, Power simulation on the actual operation in hybrid excavator, JSAE (Society of Automotive Engineers of Japan) Annual Congress, JSAE 86 (2003) 13–18.第 18 頁[8]. Y. Kanezawa, Y. Daisho, T. Kawaguchi, et al., Increasing efficiency of construction machine by hybrid system, JSAE (Society of Automotive Engineers of Japan) Annual Congress, JSAE 100 (2001) 17–20.[9]. M. Kagoshima, T. Sora, M. Komiyama, Development of hybrid power train control system for excavator, JSAE (Society ofAutomotive Engineers of Japan) Annual Congress, JSAE 86 (2003) 1–6. 第 1 頁Control strategies of power system in hybrid hydraulic excavatorQing Xiaoa,*, Qingfeng Wanga, Yanting ZhangbaThe State Key Laboratory of Fluid Power Transmission and Control, Zhejiang University, 310027 Hangzhou, ChinabCollege of mechanical and electrical engineering, China University of Petroleum, 257061 Dongying, ChinaAccepted 21 May 2007AbstractHybrid system, which has been successfully used in vehicles, is introduced to hydraulic excavators nowadays. The primary focus of this study is to investigate the control strategies of hybrid system used in hydraulic excavators. At first, the structure and working conditions of hybrid hydraulic excavators are analyzed .Based on the analyses, a control strategy named the engine constant-work-point is proposed and studied in a simulative experimental system. Then the control strategy named double-work-point is presented to overcome the limitations of the constant-work-point control strategy. The features and experimental results of the double-work-point control strategy show that the engine's efficiency and the capacitor's state of charge(SOC)cannot be optimized simultaneously. Thus a dynamic-work-point control strategy, which regulates the engine's working point dynamically, is developed to make the system work better. Experimental results show that the dynamic-work-point control strategy can improve the distribution of engine's 第 2 頁working points, restrain the capacitor's SOC and has little influence on the performance of the system.Keywords: Hybrid system; Excavator; Engine constant-work-point control strategy; Double-work-point control strategy; Dynamic-work-point control strategy1. IntroductionEnergy is consuming up and pollution is more and more serious in the world range. So research on the energy saving of construction machinery, especially hydraulic excavators, is very necessary and urgent due to their high energy consumption and bad exhaust. Traditional energy saving methods for hydraulic excavator cannot raise the effect on a large scale if there are no major technology breakthrough [1,2].It can be concluded from different working conditions of a hydraulic excavator (condition data derived from the actual work) that its load power varies periodically in a large range, thus the working condition of the engine also changes periodically and therefore cannot always remain in a high efficiency state. That's the main cause that hydraulic excavators have low fuel economy. Hybrid system, which consists of an engine and an electric motor, has the potential of improving fuel economy by operating the engine in an optimum efficiency range and it has been successfully applied in vehicles. So equipping hydraulic excavators with the hybrid system provides a new way to achieve energy savings.Recently, research on the structure, control strategy and energy management of hybrid system in hydraulic excavators has been carrying out [3–9]. Among them, the control strategy, which determines the working state of the components in the power system directly and affects the energy consumption of hydraulic excavators 第 3 頁ultimately, is one of the major concerns.This paper mainly deals with the control strategies of a hybrid system in hydraulic excavators. We present these control strategies step by step.When a hybrid system is implemented, the fluctuation of load power is absorbed by the accumulator of the power system, making the engine only output the averaged load power. Thus the control strategy of working at a constant high efficiency point can be realized for the engine with the benefit of increasing the efficiency of the engine and system.However, under the control strategy of working at a constant high efficiency point, since the chosen working power of the engine cannot be exactly the same as the average load power, the state of charge (SOC) of the accumulator will rise or drop after one work cycle. After a long time of work, the SOC will exceed its working range, and the system can work normally no longer. To overcome this limitation, we can employ a double-work point control strategy, that is, the engine works at one high power point and one low-power point in the high efficiency area. When the SOC of the accumulator exceeds the assigned upper limit, the engine switches to its low-power point; when the SOC comes to the assigned lower limit, the engine switches to its high-power point. In this way the engine's efficiency remains relatively high and the SOC of the accumulator won’t exceed its working range.Under the double-work-point control strategy, the engine will switch between two working points frequently if the assigned working range of the accumulator is narrow. This is not desirable considering the stability of the system. On the other hand, if the working range of the accumulator's SOC is set wide, the efficiency and cycle-life of the accumulator will be deteriorated. Thus a control strategy, which 第 4 頁regulates the engine's working point dynamically, has been developed to overcome this drawback in our lab. Under this control strategy, the engine's working point changes dynamically in a high efficiency range according to the accumulator's SOC, and the problems encountered in the double work-point control strategy can be avoided.The paper is organized as follows. Section 2 is devoted to the structure and working conditions of the hybrid system. Section 3 presents the engine constant-work-point control strategy. The engine double-work-point control strategy is demonstrated in section 4. The engine dynamic-work-point control strategy is addressed in section 5 together with experimental results. Finally, conclusions are provided in section 6.2. Structure and working condition of the power system2.1. Structure of the power systemFig. 1. Schematic of parallel hybrid hydraulic excavator.The structure of the power system is shown in Fig. 1. The engine and electric motor drive the hydraulic pump in a parallel hybrid style. The mechanical power of 第 5 頁the engine outputs to the hydraulic pump directly, which reduces energy conversion loss comparing with the serial hybrid system. The electric motor, which can work as a motor or generator, outputs energy together with the engine or converts the engine's redundant mechanical energy to electrical energy and stores in the capacitor.2.2. Working condition of the power systemFig. 2. Output power of the power system in digging working condition.Fig. 2 shows the normalized output power (P/Pmax) of the power system, where P is output power of the hydraulic excavator, and Pmax is the rated power of the engine. The data are derived from the actual digging work cycles of a certain hydraulic excavator. It can be seen from the figure that the output power fluctuates greatly and periodically, and the cycle time is only about 18 s. Hence a capacitor, which has a fast charge–discharge speed and long cycle-life, is used as an accumulator to balance the fast power fluctuation in the power system.第 6 頁3. Engine constant-work-point control strategy3.1. Details of the control strategyAccording to the above analyses, the load power of hydraulic excavator is regular and cyclic. The load power in one cycle can be taken as two constituent parts: the average value plus the fluctuation. So it is reasonable to employ the engine constant work- point (constant rotational speed and constant torque) control strategy for the hybrid hydraulic excavator, in which the engine works at a constant point to supply the load average power, and the fluctuating power is supplied by the electric motor-capacitor. In this way the engine can always work in its high efficiency range with high fuel economy and low emission.Under the control strategy of constant-work-point, the rotational speed of the engine is a preset value. Since the electric motor is connected with the engine coaxially, its rotational speed is the same as the engine. It can be seen from Fig. 1 that the torque of the engine is the difference of the torque of the hydraulic pump and that of the electric motor. When the load changes, we should adjust the torque of the electric motor to maintain the engine's torque constant. This can be realized by changing the revolutional slip of the electric motor via regulating its synchronous rotational speed.第 7 頁Fig. 3. Mechanical characteristic curve of electric motor.Fig. 3 shows the mechanical characteristic curve of the electric motor. Notations in the figure are the following:n Rotational speedM Torquenm Actual rotational speed of the electric motorRevolutional slip?Here nm is a constant. As shown in the figure, the mechanical characteristic curve moves up or down when the synchronous rotational speed of the electric motor changes, and the output torque of the electric motor alternates accordingly. When the synchronous rotational speed is lower than nm, becomes negative and the torque ?of the electric motor also becomes negative (the electric motor works as a generator). Otherwise and the torque of the electric motor are positive. The relationship n?between and the torque of the electric motor is decided by the motor's mechanical characteristic curve.第 8 頁Fig. 4. Control block diagram of engine torque control.Fig. 4 presents the block diagram of engine torque in the engine constant-work-point control strategy. Notations used in the figure are the following:Mei Target torque of the enginenmi Target synchronous rotational speed of the electric motornmo Actual synchronous rotational speed of the electric motor Revolutional slip of the electric motor m?Mm Output torque of the electric motorMeo Output torque of the engine Given a target torque of the engine Mei, the target synchronous rotational speed of the electric motor nmi is calculated by the control algorithm (Here the PID is chosen); the synchronous speed of the electric motor is controlled by one vector controller; the difference between nmo and nm is and the electric motor outputs ?the torque Mm according to ; then the engine outputs torque Meo to drive the m?hydraulic pump together with the electric motor.第 9 頁3.2. Experimental systemFig. 5. Schematic of the experimental system.A simulative experimental bench, illustrated in Fig. 5, was established in our lab to study the control strategies for hybrid system. A proportional relief valve was used to simulate the load pressure of hybrid system. The simulation of load flow rate was realized by alternating the displacement of the hydraulic pump. Notations in the figure are the following:pp Pressure of the hydraulic pump Q Flow rate of the hydraulic pump M1 Torque of Mot1M2 Torque of Mot2 U Voltage of the capacitor I Current of the capacitor f1 Frequency control signal of Inv1 f2 Frequency control signal of Inv2 qc Displacement control signal of the hydraulic pump 第 10 頁pc Pressure control signal of the proportional relief valveFor the convenience of control, we used a 37 kW variable frequency electric motor Mot1, which was controlled by the inverter Inv1, as the replacement of the engine in Fig. 1. A variable-frequency electric motor, Mot2, with the power of 22 kW, was controlled by the inverter Inv2. Mot1 and Mot2 were connected in parallel to drive the hydraulic pump. A set of capacitors, with the capacity of 12.5 F and maximum voltage of 400 V, was used as the accumulator of the experimental system. The main control unit of the system was composed of one industry control computer, one data acquisition card and one data control card. Appropriate sensors were used to measure pp, n, Q, M1, M2, U, I, etc. The controller collected and processed data from the sensors and output the control signals f1, f2, qc, pc to control the rotational speed of the electric motors and flow rate together with the pressure of the hydraulic system.3.3. Experimental results of the control strategyFig. 6. Pressure and flow rate of the hydraulic pump.Based on the analyses, the engine constant-work-point control strategy was 第 11 頁studied in the experimental system mentioned above. Fig. 6 shows the normalized flow rate (Q/ Qmax) and pressure (p/pmax) of the hydraulic pump in one work cycle (the data were derived from actual work cycle of a hydraulic excavator). We converted the flow rate and pressure to the corresponding signals qc and pc for the hydraulic pump and proportional relief valve in the experiment.Fig. 7. Comparison of the output power.Fig. 7 presents the comparison of normalized output power (P/Pmax) of Mot1, Mot2 and the capacitor. It can be seen that the output power of Mot1 fluctuates little during the cycle, indicating the working point of the engine is almost constant, and the output power of Mot2 is fluctuant. Fig. 7 also shows that the output power of Mot2 is always lower than that of the capacitor; the difference between them is the power conversion loss. Fig. 7 shows that the engine constant-work-point control strategy is basically feasible, but the output power of Mot1 is not exactly constant. The reason is that the algorithm of engine torque control is a simple PID and not proper enough. Improving the control algorithm is the emphasis of our next study.第 12 頁4. Engine double-work-point control strategySince the chosen working power of the engine cannot be exactly the same as the average of the load power, the SOC of he capacitor will exceed its working range after a long time of work. We further developed a control strategy in which, when the SOC exceeds its upper limit, the engine switches to a low power working point in the high efficiency range, and, when the SOC comes to its lower limit, the engine switches to a high power working point in the high efficiency range, and it is named as the engine double-work-point control strategy. The double-work-point control strategy was studied in our experimental system mentioned above. Its control method is the same as the constant-work-point control strategy, that is, the engine's torque is stabilized via adjusting the synchronous rotational speed of the electric motor. The experimental curves are shown in Fig. 8, where the engine's high-power working point is Ph, the low-power working point is Pl, P/Pmax is the normalized output power of Mot1 and S is the SOC of the capacitor. The figure illustrates working points of Mot1 switching between Pl and Ph according to the capacitor's SOC and the switch style is consistent with the above analyses, which indicates the feasibility of this control strategy. As in the engine constant-work-point control strategy, this control strategy cannot stabilize working point Pl and Ph at exactly constants either.第 13 頁Fig. 8. Experimental curves of double-work-point control strategy.It can be deduced from the experimental results that the engine will switch between the two working points frequently if the working range of the capacitor's SOC is narrow; this is not favorable for the system's stable work. If the working range of the capacitor's SOC is wide, the efficiency and working life of the capacitor will be deteriorated. Thus, a control strategy, which adjusts the engine working point dynamically, was developed to optimize the engine's working state and capacitor's SOC.5. Engine dynamic-work-point control strategy5.1. Details of the control strategyUnder this control strategy, the engine's working point is dynamically adjusted according to the capacitor's SOC after every work cycle. There are two goals of this control strategy. One is to ensure the distribution of the engine's working points in or 第 14 頁near its high efficiency range. The other is to restrain the variation range of the capacitor's SOC. The control strategy is listed below.Fig. 9. Engine's efficiency map.Step 1: Calculate the load average power and set the upper and lower limits of the engine's power. The overlapping zone between the power limits and the high efficiency area of the engine is set as the engine's working area, as shown by the dashed line area H in Fig. 9. The coordinates of Fig. 9 are normalized rotational speed (n/nmax) and the torque (M/Mmax). Step 2: Choose engine's initial working point P0(ne,Me) in the area H according to the load average power.Step 3: Set the initial capacitor's SOC S0 and the sensitivity . tS?Step 4: After i′th (i=1, 2, 3…) work cycles, if the current SOC Si and the former SOC Si ? 1 meet Eqs. (1) and (2), the system continues to work without any parameter changes; otherwise, the engine's working point is adjusted by using Eq. (3). 第 15 頁, (1)tiiSS???1, (2)ti0, (3)????????????? 011 &;,, SSKMnPn iiideieiwhere:Pi+1(ne, Me) Engine's working point after i'th work cycles Pi(ne, Me) Engine's working point after (i?1)'th work cycles Kc Adjustment coefficient when engine's power is high Kd Adjustment coefficient when engine's power is low SOC difference, equal to Si?Si ? 1 S?Step 5: Move the engine's working point into or near the H area along the power contour if necessary (as shown in Fig. 9). Step 6: Regulate the control signals of hydraulic system to drive the load according to the changed engine's working point.Step 7: As the engine's working point is regulated along the power contour, the engine's efficiency may be sacrificed to fulfill the need of the load if the adjusted hydraulic control signals are out of the control range.第 16 頁Fig. 10. Flowchart of the control strategy.The flowchart of the control strategy is shown in Fig. 10. As the engine's working point Pi+1(ne,Me) is regulated to along the power contour by ??eiMnP,1??第 17 頁adjusting its rotational speed ne and torque Me, the conditions listed below should be met:, (4)een??, (5)1?iiQwhere:Rotational speed and torque of the adjusted engine's working pointeMn?,Flow rate of the hydraulic pump before the engine's working point is 1?iQadjustedFlow rate of the hydraulic pump after the engine's working point is 1??iadjusted and , (6)11??ieiqnQ, (7)??iiwhere:qi+1 Displacement of the hydraulic pump before the engine's working point is adjustedDisplacement of the hydraulic pump after the engine's working point 1??iis adjustedFrom Eqs. (4)–(7), as the rotational speed of the engine is regulated to , the en?control torque is: eM?, (8)eeMn??And the displacement of the hydraulic pump should be:1??iq, (9)1???ieinq第 18 頁The displacement of the hydraulic pump can be regulated by controlling its stoking mechanism. The rotational speed of the engine can be adjusted by the speed regulation device, and the engine's torque is: , (10)mpeM??It can be seen from Eq. (10) that the engine's torque can be regulated by echanging the output torque of the electric motor. It has also been mentioned in mthe engine constant-work point control strategy that the change of can be mMrealized by adjusting the synchronous rotational speed of the electric motor.Thus, the control strategy can be achieved by controlling the rotational speed of the engine, the synchronous rotational speed of the electric motor and the stoking mechanism of the hydraulic pump.5.2. Experimental results of the control strategy5.2.1. Distribution of the engine's working pointsFig. 11. Distribution of engine working points without hybrid system.第 19 頁Fig. 12. Distribution of engine working points with hybrid system.Fig. 11 shows the distribution of the engine's working points when the engine drives the hydraulic system solely. As the load fluctuates, the engine's working points shift with various efficiencies. Thus the efficiency of the system cannot be very high. Fig. 12 illustrates the distribution of the engine's working points when the hydraulic system is driven by the hybrid system. Different from that shown in Fig. 11, the engine's working points concentrate in the high efficiency area, and the distribution of the working points is consistent