偷窥国产在线91,亚洲无线国产观看原创,日本精品aⅴ一区二区三区,久久九九兔免费精品6

簡介：附錄附錄1基于神經(jīng)網(wǎng)絡(luò)和遺傳算法的模糊系統(tǒng)的自動設(shè)計基于神經(jīng)網(wǎng)絡(luò)和遺傳算法的模糊系統(tǒng)的自動設(shè)計摘要摘要本文介紹了基于神經(jīng)網(wǎng)絡(luò)和遺傳算法的模糊系統(tǒng)的設(shè)計，其目的在于縮短開發(fā)時間并提高該系統(tǒng)的性能。介紹一種利用神經(jīng)網(wǎng)絡(luò)來描繪的多維非線性隸屬函數(shù)和調(diào)整隸屬函數(shù)參數(shù)的方法。還提及了基于遺傳算法的集成并自動化三個模糊系統(tǒng)的設(shè)計平臺。1前言前言模糊系統(tǒng)往往是人工手動設(shè)計。這引起了兩個問題一是由于人工手動設(shè)計是費時間的，所以開發(fā)費用很高；二是無法保證獲得最佳的解決方案。為了縮短開發(fā)時間并提高模糊系統(tǒng)的性能，有兩種獨立的途徑開發(fā)支持工具和自動設(shè)計方法。前者包括輔助模糊系統(tǒng)設(shè)計的開發(fā)環(huán)境。許多環(huán)境已具有商業(yè)用途。后者介紹了自動設(shè)計的技術(shù)。盡管自動設(shè)計不能保證獲得最優(yōu)解，他們?nèi)允强扇〉氖止ぜ记桑驗樵O(shè)計是引導(dǎo)走向和依某些標(biāo)準(zhǔn)的最優(yōu)解。有三種主要的設(shè)計決策模糊控制系統(tǒng)設(shè)計1確定模糊規(guī)則數(shù),2確定隸屬度函數(shù)的形式。3確定變化參數(shù)再者,必須作出另外兩個決定4確定輸入變量的數(shù)量5確定論證方法（1）和（2）相互協(xié)調(diào)確定如何覆蓋輸入空間。他們之間有高度的相互依賴性。（3）用以確定TSK（TAKAGISUGENOKANG）模式【1】中的線性方程式的系數(shù)，或確定隸屬度函數(shù)以及部分的MAMDANI模型【2】。4符合決定最低套相關(guān)的輸入變量,計算所需的目標(biāo)決策或控制的價值觀。像逆向消除（4）和信息標(biāo)準(zhǔn)的技術(shù)在此設(shè)計中經(jīng)常被利用。5相當(dāng)于決定使用哪一個模糊算子和解模糊化的方法。雖然由數(shù)種算法和模糊推理的方法已被提出,仍沒有選擇他們標(biāo)準(zhǔn)。5表明動態(tài)變化的推理方法,他依據(jù)這個推理環(huán)境的結(jié)果在性能和容錯性高于任何固定的推理的方法。神經(jīng)網(wǎng)絡(luò)模型（以更普遍的梯度和基于遺傳算法的神經(jīng)網(wǎng)絡(luò)最常見的梯度的基礎(chǔ)和遺傳算法被用于模糊系統(tǒng)的自動設(shè)計?；谏窠?jīng)網(wǎng)絡(luò)的方法主要是用來設(shè)計模糊隸屬度函數(shù)。這有兩種主要的方法；一直接的多維的模糊隸屬度函數(shù)的設(shè)計該方法首先通過數(shù)據(jù)庫確定規(guī)則的數(shù)目。然后通過每個簇的等級的訓(xùn)練來確定隸屬函數(shù)的形式。更多細(xì)節(jié)將在第二章給出。二間接的多維的模糊隸屬度函數(shù)的設(shè)計這種方法通過結(jié)合一維模糊隸屬函數(shù)構(gòu)建多維的模糊隸屬度函數(shù)。隸屬度函數(shù)梯度技術(shù)被用于調(diào)節(jié)試圖減少模糊系統(tǒng)的期望產(chǎn)量和實際生產(chǎn)所需的產(chǎn)出總量的誤差。第一種方法的優(yōu)點在于它可以直接產(chǎn)生非線性多維的模糊隸屬度函數(shù)；沒有必要通過結(jié)合一維模糊隸屬函數(shù)構(gòu)建多維的模糊隸屬度函數(shù)。第二種方法的優(yōu)點在于可通過監(jiān)測模糊系統(tǒng)的最后性能來調(diào)整。這兩種方法都將在第二章介紹。許多基于遺傳算法的方法與方法二在本質(zhì)上一樣；一維隸屬函數(shù)的形式利用遺傳算圖1模糊劃分（A）常規(guī)（B）期望圖2NNDRIVEN結(jié)構(gòu)模糊推理實例2222調(diào)整參數(shù)的模糊系統(tǒng)調(diào)整參數(shù)的模糊系統(tǒng)這個定義隸屬度函數(shù)形式的參數(shù)來減少模糊系統(tǒng)輸出和監(jiān)督的數(shù)據(jù)之間的誤差。兩種方法用于修改這些參數(shù)摘要現(xiàn)有基于梯度方法和遺傳算法。遺傳算法的方法將在下一章節(jié)講述,基于梯度的方法將在這部分解釋。這個基于梯度的方法的程序是1決定如何確定的隸屬度函數(shù)的形式2利用梯度方法調(diào)整降低模糊系統(tǒng)的實際輸出與期望輸出的參數(shù)，通常最速下降。隸屬函數(shù)的中心的位置和寬度通常用來定義參數(shù)的形狀。ICHIHASHIETAL6ANDNOMURAETAL7,8,HORIKAWAETAL910,ICHIHASHIETALLLANDWANGETAL12,JANG1314已經(jīng)分別用三角形,結(jié)合SIGMOIDAL、高斯,鐘型隸屬度函數(shù)。他們利用最速下降法來調(diào)整模糊隸屬函數(shù)參數(shù)。圖3神經(jīng)網(wǎng)絡(luò)調(diào)整模糊系統(tǒng)的參數(shù)圖4調(diào)整模糊系統(tǒng)的神經(jīng)網(wǎng)絡(luò)圖3顯示了此方法和同構(gòu)于圖4圖中的UIJ在ITH規(guī)則下輸入模糊隸屬函數(shù)的參數(shù)XJ,而它實際上是代表一個描述隸屬度函數(shù)的形式的參數(shù)向量。也就是說，這個方法使模糊系統(tǒng)作為神經(jīng)網(wǎng)絡(luò)的模糊隸屬度函數(shù)和通過節(jié)點執(zhí)行重量和規(guī)則一樣。任何網(wǎng)絡(luò)學(xué)習(xí)算法,例如反向傳播算法，可以用來設(shè)計這種結(jié)構(gòu)。3遺傳算法方法遺傳算法方法31遺傳算法與模糊控制遺傳算法與模糊控制遺傳算法是進行優(yōu)化、生物激勵的技術(shù)，他的運行用二進制表示，并進行繁殖,交叉和變異。繁殖后代的權(quán)利是通過應(yīng)用程序提供的一種健身價值。遺傳算法吸引人是因為他們不需要存在的衍生物,他們的同時搜索的魯棒性很強，并能避免陷入局部最小。SEVERM的論文提出了利用自動遺傳算法的模糊系統(tǒng)的設(shè)計方法。大量的工作主要集中在調(diào)整的模糊隸屬度函數(shù)1725。其他的方法使用遺傳算法來確定模糊規(guī)則數(shù)18,26。在26中,通過專家制定了一系列規(guī)則，并且遺傳算法找到他們的最佳的組合。在18,卡爾已經(jīng)開發(fā)出一種方法用于測定模糊隸屬度函數(shù)和模糊規(guī)則數(shù)。在這篇文章中,卡爾的方法首先用遺傳算法按照預(yù)先定義的規(guī)則庫確定規(guī)則的數(shù)目。這個階段后,利用遺傳

簡介：出處出處MUTOHN,AKASHIHELECTRICANDMECHANICALBRAKECOOPERATIVECONTROLMETHODFORFRIDEVSUNDERVARIOUSSEVEREROADCONDITIONSC//IECON201137THANNUALCONFERENCEONIEEEINDUSTRIALELECTRONICSSOCIETYIEEE,201145704576中文中文6955字在各種嚴(yán)重路況下FRID電動汽車的電氣和機械聯(lián)合制動MUTOHN,AKASHIH摘要本文描述了一種適合前后車輪獨立驅(qū)動的電動汽車FRID電動汽車在摩擦系數(shù)M較低的道路條件下，如結(jié)冰路面和摩擦不均勻路面，電氣和機械聯(lián)合制動的方法。聯(lián)合制動方法的提出是基于分析車輛在摩擦不均勻道路上制動是最危險的操作行為的結(jié)果。在提出的制動方法中,控制制動力以抑制發(fā)生在左、右車輪之間除轉(zhuǎn)向所必須的側(cè)向力外的側(cè)向力力矩不平衡情況的發(fā)生。這種制動方式考慮了電氣和機械剎車在響應(yīng)速度之間的最大的區(qū)別。這里通過模擬和實驗證實，在滑移率值不同的各種嚴(yán)重路況下使用機械和電氣聯(lián)合制動的制動方法都可獲得穩(wěn)定的制動性能。圖11FRIDEV安全性和運行性能兼容的驅(qū)動結(jié)構(gòu)圖12在當(dāng)前的防抱死制動系統(tǒng)ABS中操作域有效現(xiàn)在,研究汽車在各種道路條件下的制動控制方法是非常重要的14,特別是在低摩擦道路條件下實現(xiàn)安全制動是很困難的。圖12顯示了在當(dāng)前的防抱死制動系統(tǒng)ABS中操作域有效15。目前ABS系統(tǒng)存在在制動器力量很小的領(lǐng)域不運作的問題。這意味著在低摩擦路面如結(jié)冰的路面強烈踩下制動踏板時極易發(fā)生制動跑偏。電動剎車在這樣的低摩擦路面上能有效的準(zhǔn)確的控制制動力。目前對兩輪驅(qū)動式和四輪驅(qū)動式電動汽車的ABS系統(tǒng)已進行了許多研究1618。自獨立控制每個車輪的制動力后,出現(xiàn)了在摩擦不均勻路面上車輛可操縱性惡化的問題。此外,因為很難正確補償四個輪子減速時產(chǎn)生的負(fù)荷運動,所以減速期間車輛可能會變得不穩(wěn)定。在運行條件下,可以用相對較少的制

簡介：中文中文3560字出處出處CONTROL,AUTOMATION,ROBOTICSANDVISION,2008ICARCV200810THINTERNATIONALCONFERENCEONIEEE,200816251629畢業(yè)設(shè)計（論文）外文文獻翻譯畢業(yè)設(shè)計（論文）外文文獻翻譯畢業(yè)設(shè)計（論文）題目翻譯題目翻譯題目化工廢水自動化處理系統(tǒng)的控制器設(shè)計與分析化工廢水自動化處理系統(tǒng)的控制器設(shè)計與分析系自動控制系自動控制系專業(yè)自動化自動化姓名班級學(xué)號指導(dǎo)教師指導(dǎo)教師學(xué)探針。這種方法需持續(xù)監(jiān)控，導(dǎo)致了高成本，大量勞動力和低效率。此外工廠中的一些地方對人來說是無法達到的，同時有害化學(xué)品的介入對人體健康也是有害的。本文完整地提出了一種控制化工廢水處理廠工藝過程的全自動解決方案。水流量，鍋爐加熱溫度和冷卻水的溫度流量都是依靠化工廢水的輸入輸出濃度和加熱溫度來被調(diào)控。（在串級控制器的情況下）III數(shù)學(xué)模型工廠的數(shù)學(xué)模型是基于同步獲得的隨機輸入輸出信號的特點。這個構(gòu)想已經(jīng)被應(yīng)用到定常線性動態(tài)化工處理廠中，其模型是由加權(quán)函數(shù)和傳遞函數(shù)描述的。該工廠的加權(quán)函數(shù)是由采用積分WIENERHOPF方程求得輸入的自相關(guān)函數(shù)和輸入輸出的互相關(guān)函數(shù)確定的。傳遞函數(shù)涉及到加權(quán)函數(shù)的拉普拉斯變換。1G432231432231PASASASABSBSBSBS???????）（在參考1中實現(xiàn)的方法對于廣大工業(yè)用戶去推導(dǎo)化工廢水處理廠的傳遞函數(shù)是簡單，有效和可行的，因為它采用了標(biāo)準(zhǔn)的數(shù)學(xué)方法并且通過含有時域和頻域動態(tài)特性的確定的平均數(shù)學(xué)模型描述了整個系統(tǒng)。該方法還可以構(gòu)建一個粗略的結(jié)構(gòu)，這對于一個真實的化工廢水處理廠的綜合和分析是很有必要的。本文提出的控制相關(guān)參數(shù)如輸出濃度，反應(yīng)器溫度，流量，冷卻水溫度和處理廠流量的設(shè)計的輸入輸出關(guān)系是由傳遞函數(shù)決定的。在設(shè)計中，先進的過程控制算法和控制技術(shù)被采用，并且通過實時工業(yè)數(shù)據(jù)表可以進行分析工作。IV系統(tǒng)設(shè)置化工廢水處理廠包括嵌有流量傳感器的輸入管道，在凈化裝置前端放置一個密度傳感器，PID、串級、前饋控制器和反應(yīng)器一起構(gòu)成了凈化裝置。輸出管道也設(shè)置了密度計和流量傳感器，整個自動化工廠都是由PLC控制的。此外，輸出單元還包括了一個廢棄物分離器和一個固體干燥器，他們的用途是把廢棄物從廢水中分離出來并且加熱干燥用于進一步的工業(yè)應(yīng)用中。圖1自動化化工廢水處理廠設(shè)計布局附控制器和傳感器S1,S2,S3,S4流量傳感器

簡介：中文中文1萬字萬字出處出處SHANGHC,CHENYP,YUWY,ETALONLINEAUTODETECTIONMETHODANDSYSTEMOFPRESSWORKQUALITYJINTERNATIONALJOURNALOFADVANCEDMANUFACTURINGTECHNOLOGY,2007,3378756765本科生畢業(yè)設(shè)計外文文獻翻譯學(xué)生姓名指導(dǎo)教師專業(yè)班級學(xué)號學(xué)院11引言通過設(shè)計和加工控制的整合信息，機器視覺系統(tǒng)能夠獲得快速的獲得廣闊的信息。因此，機器視覺技術(shù)作為檢查方法已經(jīng)很快速的進行了發(fā)展，并已在各行業(yè)檢驗過程的應(yīng)用，例如作為后粘接，鋼材表面的缺陷、織物檢查，故障的水瓶，硬幣檢測等。在線檢測印刷品在行業(yè)檢測中的應(yīng)用機器視覺域上也被引進。自定義檢測印刷品包括印刷品裸眼后復(fù)查手冊觀察網(wǎng)上，通過監(jiān)視器監(jiān)視進程和隨機檢查等，并且具有的缺點包括毛坯料，視疲勞的嚴(yán)重浪費，抽查無法保證整體質(zhì)量，等等。隨著印刷技術(shù)的發(fā)展，在印刷速度的提高，及數(shù)據(jù)的復(fù)雜性和信息，印刷質(zhì)量控制的手動模式已經(jīng)成為不適合的實踐需求。為了保證印刷質(zhì)量，保證印刷品的整體，提高了生產(chǎn)速度，這是實時在線質(zhì)量檢測印刷品中必須進行的。近年來，高速CCD廣泛應(yīng)用于工業(yè)相機技術(shù)和高性能的微電腦,不僅能提供更好的硬件,也使實時在線質(zhì)量檢測1得以實現(xiàn)。該高實時性的算法和研究在線自動檢測系統(tǒng)的研究，促進人員按VOL工作要求來應(yīng)付顧客需求。這些機器的視覺驅(qū)動力系統(tǒng)減少了打印浪費的20％40％，而系統(tǒng)則允許印刷機以高速操作，以滿足生產(chǎn)率要求的必要，以確保持久的客戶滿意度。應(yīng)用特殊機器視覺系統(tǒng)或視頻檢測系統(tǒng)開發(fā)的印刷行業(yè)，像所有的機器視覺產(chǎn)品一樣，從拍攝的圖像、面積攝像機或線性陣列。機器視覺系統(tǒng)在印刷行業(yè)包括兩個版本被動和主動。被動系統(tǒng)自動完成整個檢查過程并刪除操作者在一些情況下循環(huán)。他們可以識別隨機和重復(fù)的打印缺陷，并可以用于從壞的物質(zhì)提取那種良好的，它們可以100％使用一個高的網(wǎng)絡(luò)和印刷圖案的檢查分辨率區(qū)域相機或線陣相機。本文研究的是一種被動的系統(tǒng)。主動檢查系統(tǒng)需要操作者的介入。今天的這些系統(tǒng)大多數(shù)上執(zhí)行一些機器視覺樣，用更智能的系統(tǒng)、更少運營商參與的基本觀察和決策。最起碼，在這些情況下，圖像可以由系統(tǒng)被放大以使其更容易為一個操作員查看印刷圖案，并顯示在分割屏幕到所采樣的圖象進行比較的基準(zhǔn)引用的圖像。許多無源系統(tǒng)能夠進行自動比較監(jiān)測的色彩，評估條形碼質(zhì)量，監(jiān)視和正確的注冊，并檢查打印缺陷。這些高端系統(tǒng)錯誤時，提醒操作者發(fā)現(xiàn)并顯示在網(wǎng)絡(luò)上有缺陷的位置條件。在某些情況下，這些系統(tǒng)可以啟動相關(guān)的控制新聞自動糾正措施上的部分。圖像處理軟件自動著眼于在網(wǎng)絡(luò)上的特定位置和樣本顏色進行比較，以一個標(biāo)準(zhǔn)的顏色，從存儲視顏色到顏色套準(zhǔn)。在許多系統(tǒng)中，這是通過監(jiān)測在幅材的邊緣或在該模式內(nèi)專用標(biāo)志完成印刷。一些系統(tǒng)現(xiàn)在可以查看打印本身在這一塊

簡介：球形研磨和拋光注塑模具鋼的自動化表面精加工工藝球形研磨和拋光注塑模具鋼的自動化表面精加工工藝FANGJUNGSHIOUCHAOCHANGACHENWENTULI收件日期2004年3月30日/接受日期2004年7月5日/發(fā)表時間05年3月30號?施普林格出版社倫敦有限公司2005摘要摘要本研究討論在數(shù)控加工中心注塑模具鋼PDS5在自由曲面下進行自動化球形研磨和拋光球的表面處理工藝的可行性。研磨工具持有人的設(shè)計和制造已經(jīng)完成了這項研究。在加工中心中，表面的最佳磨削參數(shù)采用田口直交法來進行塑料注射成型鋼PDS5而確定。塑料注塑模具鋼PDS5表面最佳磨削參數(shù)是，一種PA的氧化鋁磨削材料組合，以18000RPM的速度，20ΜM的磨削深度，以及50毫米/分鐘的進給速度磨削。試樣的表面粗糙度RA可以通過使用最佳的表面磨削參數(shù)來從160微米大約提高至035微米。表面粗糙度RA還可通過使用最佳拋光參數(shù)的球拋光這一過程進一步改善至約0343微米至006微米。應(yīng)用表面打磨和拋光最佳參數(shù)，依次細(xì)研磨自由曲面模仁，自由曲面上測試區(qū)的表面粗糙度RA部分可提高到約215微米至007微米。關(guān)鍵詞關(guān)鍵詞自動化表面精加工球打磨過程表面粗糙度磨削工藝田口方法11簡介簡介塑料是重要的工程材料，由于其特性，如耐腐蝕，耐化學(xué)品，密度低，易于制造，并已在工業(yè)應(yīng)用中越來越多地取代金屬部件。注塑成型是一種重要的塑料產(chǎn)品成型工藝，塑料模具表面光潔度是一個直接影響塑料產(chǎn)品外觀的必要條件。如磨削，拋光和研磨這樣的整理程序常用來改善表面光潔度。研磨工具砂輪的裝入已經(jīng)廣泛使用的傳統(tǒng)模具，模具加工等行業(yè)。為了自動化表面精加工進程，安裝了磨削工具的幾何磨具在（1）中引入。在自動化表面精加工系統(tǒng)中，球形研磨的球形研磨工具的加工進程模型在（2）中闡述。磨削速度，切削深度，進給速度，研磨材料，磨料，料度等砂輪特性都為球面磨削過程的影響參數(shù)，如圖1所示。注塑模具鋼的最佳球面磨削參數(shù)尚未在文獻中調(diào)查發(fā)現(xiàn)。近年來，正在開展一些研究來確定球擠光過程的最佳參數(shù)（圖2）。據(jù)說使用碳化鎢球或滾子可減小工件表面的塑性變形，因而改善表面粗糙度，表面硬度和抗疲勞性能36。打磨過程是由加工中心3，4和車床5，6完成。對表面粗糙度有顯著影響的主要拋光參數(shù)是滾珠或滾子的材料，打磨力，進給速度，拋光速度，潤圖2球擠光過程示意圖BURNISHINGDIRECTION磨削方向ORIGINALSURFACETEXTURE原有的表面紋理FEED進給BURNISHEDSURFACETEXTURE磨削后表面紋理DEPTHOFPENETRATION穿透深度ORIGINALHEIGHT原有深度BURNISHINGBALL拋光球WORKPIECE工件FINISHEDHEIGHT加工后深度

簡介：畢業(yè)論文（設(shè)計）畢業(yè)論文（設(shè)計）外文翻譯題目電氣工程畢業(yè)生使用MATLAB最優(yōu)控制的過程系部名稱系部名稱專業(yè)班級專業(yè)班級學(xué)生姓名學(xué)生姓名學(xué)號號指導(dǎo)教師指導(dǎo)教師教師職稱教師職稱講師20年月日使用這些程序設(shè)計包的三個缺點是使用這些程序設(shè)計包計算機容易受影響，而且還要求學(xué)生或者老師學(xué)會如何使用這些計算機輔助程序設(shè)計包，并保證這些包括基準(zhǔn)線課程的設(shè)計包作為必修課程的一部分，更多詳細(xì)關(guān)于它的缺點的討論可以參閱我們前部分章節(jié)的內(nèi)容。賓州法尼亞大學(xué)大峽谷分校的最有控制系統(tǒng)賓州法尼亞大學(xué)大峽谷校園是賓州法尼亞大學(xué)十八個校園的其中之一，是在費城區(qū)域的一個研究生中心，工程師們在那里工作來滿足教育的需求，利用各種各樣的仿真設(shè)計包幾乎所有的學(xué)生都變成了工程師。這些課程的目的是讓學(xué)生的數(shù)學(xué)工具參數(shù)化，動態(tài)最優(yōu)化，并且用于設(shè)計表現(xiàn)最好的動力系統(tǒng)。非線性系統(tǒng)課程包含下面的課題1、靜態(tài)最佳化2、離散系統(tǒng)的時間最優(yōu)控制21線性二次調(diào)節(jié)器22次優(yōu)反饋的穩(wěn)態(tài)閉環(huán)控制系統(tǒng)23跟蹤問題24最終狀態(tài)固定的調(diào)節(jié)閥與功能3、最優(yōu)控制的連續(xù)系統(tǒng)31線性二次調(diào)節(jié)器32次優(yōu)反饋的穩(wěn)態(tài)閉環(huán)控制系統(tǒng)33跟蹤問題34最終狀態(tài)固定的調(diào)節(jié)閥與功能35最終時間釋放問題36約束輸入問題4、動態(tài)規(guī)劃41離散時間學(xué)生每周的作業(yè)包括計算機的仿真和使用，考試部分也包括用計算機仿真的作業(yè)，學(xué)生可以使用學(xué)生版本的MATLAB。阿肯色州大學(xué)最優(yōu)控制系統(tǒng)本課程提供給本科生和研究所的學(xué)生，讓他們?yōu)檠芯拷?jīng)典的狀態(tài)空間控制技術(shù)做好充分準(zhǔn)備。這些課程的目的是讓學(xué)生的數(shù)學(xué)工具參數(shù)化，動態(tài)最優(yōu)化，而且用于設(shè)計最好的動力系統(tǒng)。為實現(xiàn)這一目標(biāo)，作者（EYAZ）從1986年開始更新課程，從而下面

簡介：基于標(biāo)準(zhǔn)化數(shù)字處理信息的PLC軟件自動生成摘要摘要為滿足汽車產(chǎn)業(yè)中的諸多挑戰(zhàn)，研究新的集成的生產(chǎn)規(guī)劃與提速生產(chǎn)過程的方法是必要的。因此,鑒于汽車產(chǎn)業(yè)實體店需要的特殊關(guān)注，一個基于標(biāo)準(zhǔn)化數(shù)字處理信息的全自動PLC軟件生成的概念被引入了。PLC軟件生成的基礎(chǔ)是一個數(shù)字化處理規(guī)劃，它用圖形化的方法描述生產(chǎn)步驟。由于這種PLC軟件生成的特殊需求，引入了新的進程元素及屬性。所以進程規(guī)劃包含了PLC程序所有逐步專業(yè)化的必要信息。本文還給出一種規(guī)格擴展帶有特殊資源PLC功能模塊的數(shù)字處理信息。關(guān)鍵詞關(guān)鍵詞PLC；軟件自動生成；進程規(guī)劃；IEC61131–31．前言由于主要市場劃分中競爭的日益激烈，汽車制造商忙于通過大量的產(chǎn)品來增強競爭力。相應(yīng)的，創(chuàng)新和模型的周期不斷減小。激烈的競爭狀況不可避免地導(dǎo)致在整個生產(chǎn)規(guī)劃和提速生產(chǎn)過程的加度復(fù)雜化。另外，未來生產(chǎn)工程項目成熟的機會也大大減小。如今,生產(chǎn)規(guī)劃階段的活動機械設(shè)計,電子設(shè)計,控制工程,幾乎都要按順序完成，特別是機械設(shè)計沒有差不多完成之前不進行控制工程的設(shè)計。除此之外,在機械設(shè)計和控制工程間存在很大的差距，尤其在使用不同的工作方法、工具、使用不同的具體部門的條款時這種現(xiàn)象格外明顯。具體部門的差異不可避免的造成信息的丟失，重復(fù)的手動輸入，額外的迭代循環(huán)以及對PLC項目高度的時間壓力。為了應(yīng)對這些挑戰(zhàn)，新的規(guī)劃方法必須鼓勵同時工程。本文中介紹的全自動PLC軟件生成的新方法便能夠為同時工程做出重大貢獻。作為概述，該文以一個看待這個問題重要而有代表性的方法的簡單描述作為開始，基于現(xiàn)存方法明確的優(yōu)點和弊病，給出了對于一個全自動PLC軟件生成的新的綜合方法的要求。簡短的介紹了數(shù)字化工廠的概念之后，介紹了能夠滿足前面所提要求的新方法，包括了從數(shù)字化進程規(guī)劃到一個符合IEC61131–3規(guī)則的PLC程序等幾個步式下完成的。全自動PLC代碼生成的基礎(chǔ)是電子元件傳感器，致動器。這些對象都化做三維圖像被嵌入到單元模型里。在下一步中通過與用戶預(yù)先互動（可視化程序），把這些獨立的對象邏輯連接到布爾操作器上，借助于一個特別發(fā)達的IL專有的編譯器，這些信息就從虛擬現(xiàn)實工具轉(zhuǎn)變成了機器可識別的模式，然后生成的PLC代碼被下載到一個PLC編程工具。OSMERS所開發(fā)的這個軟件的專用傳輸?shù)母拍羁梢杂商摂M現(xiàn)實工具VRTSUPERSCAPE有限公司以及PLC編程工具STEP7西門子來實現(xiàn)。2．4．BMBF的項目MODALEBMBF項目MODALE關(guān)注的焦點在于公司范圍內(nèi)，參與在自動化生產(chǎn)單元規(guī)劃過程的各不同部門之間的數(shù)據(jù)交換。為此,開發(fā)了面向本體的相關(guān)模型嵌入在一種技術(shù)基礎(chǔ)設(shè)施中執(zhí)行中央數(shù)據(jù)撥號功能?；谶@些技術(shù)條件，數(shù)字規(guī)劃工具的數(shù)據(jù)在語義橋梁的幫助下輸出和轉(zhuǎn)移到面向本體的相關(guān)模型上。此處數(shù)據(jù)發(fā)生轉(zhuǎn)換，生成的數(shù)據(jù)輸入到各自的目標(biāo)應(yīng)用。因此可以做到在不同的IT工具之間交換數(shù)據(jù)規(guī)劃信息而不會丟失。使用初等實例論證了IT基礎(chǔ)設(shè)施的原型轉(zhuǎn)換以及主要可操作性的驗證。一個實際場景著目于一個與IEC611313相關(guān)的PLC序列運行功能表的自動生成?；灸Ｐ褪且环N用PLC專用信息例如轉(zhuǎn)換生成生產(chǎn)規(guī)劃與改善規(guī)劃的數(shù)字過程開發(fā)計劃。該模型使用一種特殊擴展的RDF出口RDF資源描述框架，參考相關(guān)模型轉(zhuǎn)化并最終以SFC格式輸入到目標(biāo)系統(tǒng)。數(shù)字過程開發(fā)計劃由DELMIA公司的數(shù)字過程規(guī)劃工具DELMIA過程工程師DPE生成。使用OPENPCS作為PLC編程工具，有公司的信息小組開發(fā)并分派。2．5．過程仿真調(diào)試的PLC軟件生成相比于前面部分提到的方法,IT工具過程仿真模擬代表了TECNOMATICS與西門子合作于2002年開發(fā)的一種商業(yè)軟件工具。PLC軟件自動化生成的基礎(chǔ)是一個數(shù)字三維單元模型和本單元的數(shù)字化進程描述。仿照MODALE項目延伸的方法，數(shù)字化進程描述隨控制技術(shù)信息增強，例如，輸入/輸出信號。通過手動添加所需參數(shù)或?qū)①Y源夾、閥門、機器人等與數(shù)字進程模型中相對應(yīng)的操作連接均可增強描述。通

簡介：1ABSTRACTTHECOMBUSTIONSYSTEMOFCIRCULATINGFLUIDIZEDBEDBOILERCFBBISACOMPLEXOBJECTWITHMULTIVARIABLE,LARGEDELAY,TIGHTLYCOUPLING,NONLINEARANDSLOWTIMEVARYINGITISDIFFICULTTOBUILDTHEPRECISEMATHEMATICMODELOFTHEOBJECTANDTOACCURATELYCONTROLTHEOBJECTWITHTHETRADITIONALCONTROLMETHODSINTHISPAPER,THEOBJECTISDYNAMICALLYDECOUPLEDBYAFEEDFORWARDCOMPENSATORTHEN,THEOBJECTISRESPECTIVELYCONTROLLEDBYTHREECONTROLLERS,PIDCONTROLLER,FUZZYCONTROLLERANDSELFTUNINGPARAMETERFUZZY–PIDCONTROLLERTHESIMULATIONEXPERIMENTSARECARRIEDOUTTHROUGHCONTRASTWITHFORMERCONTROLLERSINMATLABSIMULATIONENVIRONMENTTHESIMULATIONRESULTSSHOWTHATTHESELFTUNINGPARAMETERFUZZYPIDCONTROLLERISSUPERIORTOTHEGENERALPIDCONTROLLERANDTHEGENERALFUZZYCONTROLLERINSPEEDINESS,STABILITY,ADAPTABILITY,ROBUSTNESSANDABILITYOFANTIDISTURBANCEINDEXTERMSCIRCULATINGFLUIDIZEDBEDBOILERDECOUPLINGFUZZYCONTROLMULTIVARIABLESELFTUNINGPARAMETERFUZZYPIDCONTROLI．INTRODUCTIONHECIRCULATINGFLUIDIZEDBEDBOILERCFBBHASBEENWIDELYUSEDASACLEANCOALCOMBUSTIONTECHNIQUEFORITSHIGHCOMBUSTIONEFFICIENCY,WIDEADAPTABILITYTOCOALRANKS,LOADADJUSTINGPERFORMANCEANDLOWPOLLUTION14HOWEVER,DUETOITSSPECIALSTRUCTUREANDTHECOMPLEXITYOFCOMBUSTIONMECHANISM,THECOMBUSTIONPROCESSHASCOMPLEXFEATURES,SUCHASHIGHLYNONLINEAR,TIMEVARYING,LARGEDELAYANDMULTIVARIABLEDECOUPLING,ETCITISVERYDIFFICULTTOESTABLISHITSPRECISEMATHEMATICALMODEL58,ANDTOCONTROLTHEOBJECTWITHTHETRADITIONALCONTROLMETHODSATPRESENT,THECOMMONCONTROLMETHODISTHATTHEMAINSTEAMPRESSUREISFOCALLYCONTROLLED,MEANWHILE,THEPRIMARYAIRFLOWISADJUSTEDACCORDINGTOTHEBESTAIRCOALRATIO,ANDTHEBEDTEMPERATUREISCONTROLLEDINTHEREQUIREDRANGE9THEMETHODCAN’TMAINTAINTHEBEDTEMPERATUREINTHEBESTRANGEWHILEKEEPINGTHEMAINSTEAMPRESSUREINTHISPAPER,ONTHEBASISOFDECOUPLINGMODELOFTHEMAINSTEAMPRESSUREANDTHEBEDTEMPERATURE10,THESELFTUNINGFUZZYPIDCONTROLLER1118WHICHHASBETTERADAPTABILITYANDTHISWORKWASSUPPORTEDBYLEADINGACADEMICDISCIPLINEPROJECTOFSHANGHAIMUNICIPALEDUCATIONCOMMISSIONPROJECTNUMBERJ51301ANDNATURESCIENCEKEYFOUNDATIONOFSHANGHAIMUNICIPALEDUCATIONCOMMISSIONPROJECTNUMBER06ZZ69ALLOFTHEAUTHORSAREWITHCOLLEGEOFELECTRICPOWERANDAUTOMATION,SHANGHAIUNIVERSITYOFELECTRICPOWER,YANGPUDISTRICT,SHANGHAI,200090,CHINAEMAILCHENGQIMINGSINACOMBETTERROBUSTNESS,ISUSEDTOCONTROLTHESTEAMPRESSUREANDTHEBEDTEMPERATURETOGETTHEBETTERCONTROLEFFECTSII．THECHARACTERISTICSOFCFBBANDTHESTRUCTUREOFTHEDECOUPLEDCONTROLSYSTEMOFCFBBTHEKEYREASONSTHATCFBBCOMBUSTIONSYSTEMISDIFFICULTTOCONTROL,ARETHESTRONGCOUPLINGRELATIONSBETWEENMULTIINPUTSCOAL,PRIMARYAIR,SECONDARYAIR,DRAWINGWIND,RECYCLEMATERIALANDMULTIOUTPUTSBEDTEMPERATURE,MAINSTEAMPRESSURE,THEFURNACENEGATIVEPRESSURE,OXYGENCONTENT,THEMOSTIMPORTANTRELATIONINTHESECOUPLINGRELATIONSISTHECOUPLINGRELATIONBETWEENBEDTEMPERATUREANDMAINSTEAMPRESSURE56INCHINA,CFBBAREUSUALLYDESIGNEDWITHOUTEXTERNALHEATEXCHANGERTOENSURESIMPLESTRUCTUREANDLOWCOSTTHEMAINSTEAMPRESSUREANDTHEBEDTEMPERATUREWITHSTRONGLYCOUPLEDRELATIONARECONTROLLEDBYREGULATINGTHECOALAMOUNTANDTHEPRIMARYAIRAMOUNTTHISCONTROLMETHODISWIDELYUSEDTOTHEACTUALCONTROLOFCFBBCOMBUSTIONSYSTEMINCHINA12THEREFORE,INTHISPAPER,THEBEDTEMPERATUREISCONTROLLEDBYTHEPRIMARYAIRAMOUNT,ANDTHEMAINSTEAMPRESSUREISREGULATEDBYTHECOALAMOUNTINTHISPAPER,THEDOMESTIC75T/HCFBBISSELECTEDASTHEOBJECTOFSIMULATIONEXPERIMENTSTHETRANSFERFUNCTIONMATRIXOFTHESYSTEMATTHELOADRANGEOF70TO100ISGOTTEN16???????????????????????????????????????12602230212221121101160036012600066011638767811803860QBSESSESQBSGSGSGSGPTSSB1WHERETB,P0ARERESPECTIVELYBEDTEMPERATUREANDMAINSTEAMPRESSUREB,Q1ARERESPECTIVELYCOALAMOUNTANDPRIMARYAIRAMOUNTG11,G12G21,G22ARERESPECTIVELYINPUTOUTPUTTRANSFERFUNCTIONSOFBTB,BP0,Q1TBANDQ1P0FROMEQ1TOSEE,THATTIMEDELAYSEXISTINBOTHCOALMAINSTEAMPRESSURELOOPANDCOALBEDTEMPERATURELOOPANDTHEREARESERIOUSLYCOUPLINGRELATIONSINCFBBSYSTEMTHEREFORE,DYNAMICFEEDFORWARDCOMPENSATIONISNEEDEDFORDYNAMICDECOUPLINGSYSTEMTOCONTROLTHESYSTEMBETTERDYNAMICFEEDFORWARDCOMPENSATIONISCOMMONLYUSEDFORSYSTEMDECOUPLING10FIG1ISTHESTRUCTUREOFTHERESEARCHONMULTIVARIABLEDECOUPLINGCONTROLSYSTEMFORCOMBUSTIONSYSTEMOFCIRCULATINGFLUIDIZEDBEDBOILERQIMINGCHENG,RUIQINGGUO,ANDXUFENGDUT3TRIALANDERRORMETHODIV．THEDESIGNOFSELFTUNINGFUZZYPIDCONTROLLERFORCOMBUSTIONCONTROLSYSTEMOFCFBBATHESTRUCTUREOFCONTROLSYSTEMTHEPIDCONTROLLERHASMANYADVANTAGES,SUCHASSIMPLESTRUCTURE,MATURETHEORETICALBASIS,WIDEAPPLICABILITY,CONVENIENTPARAMETERTUNING,MUCHENGINEERINGAPPLICATIONTHEREFORE,THEPIDCONTROLLEROCCUPIESADOMINANTPOSITIONINTHEACTUALCONTROLSYSTEMBUTTHELINEARCHARACTERISTICSOFTHECONVENTIONALPIDCONTROLLERWITHFIXEDCONTROLPARAMETERSHAVEGOODCONTROLPERFORMANCEONLYWHENWORKINGNEAROPERATINGPOINTWHENTHESYSTEMISFARTHEROUTOFTHEOPERATINGPOINT,NONLINEARCONTROLCHARACTERISTICSOFTHEOBJECTISDIFFICULTTOMAINTAINTHEDYNAMICQUALITYOFPIDCONTROLTHEREFORE,FUZZYREASONINGISINTRODUCEDTOSOLVEDTHEPROBLEM,THEPARAMETERSOFPIDCONTROLLERBASEDONTHEINITIALPIDCONTROLPARAMETERSARECORRECTEDWITHADDINGFUZZYREASONINGTOIMPROVETHESYSTEMDYNAMICPERFORMANCEINSELFTUNINGFUZZYPIDCONTROLLER1618,THECONDITIONSANDTHEOPERATIONSOFCONTROLRULESAREEXPRESSEDBYFUZZYSETONTHEBASISOFPIDCONTROL,ANDTHESEFUZZYCONTROLRULESASWELLASOTHERINFORMATIONARESTOREDINTOCOMPUTERKNOWLEDGEBASES,THEN,ACCORDINGTOACTUALRESPONSEOFTHECONTROLSYSTEM,FUZZYREASONINGISCARRIEDOUTBYCOMPUTERTOACHIEVEBESTADJUSTMENTOFPIDCONTROLTHESTRUCTUREOFSELFTUNINGPARAMETERFUZZYPIDCONTROLSYSTEMISSHOWNINFIG5FIRSTLY,THESELFTUNINGFUZZYPIDCONTROLLERISDESIGNEDTOFINDTHEFUZZYRELATIONSBETWEENTHETHREECONTROLPARAMETERSNAMELY,PROPORTIONALCOEFFICIENTKP,DIFFERENTIALCOEFFICIENTKDANDINTEGRALCOEFFICIENTKIANDTHETWOSYSTEMVARIABLESTHATIS,ERRORE,ERRORDEVIATIONECTHEN,ΔKP,ΔKI,ΔKD,THECHANGESOFKP,KI,KDPIDCONTROLPARAMETERSAREONLINEAMENDEDONFUZZYTHEORYWITHMEASURINGEANDECDURINGTHESYSTEMOPERATIONFINALLY,THECONTROLSYSTEMHASGOODDYNAMICANDSTATICPERFORMANCETHEDIGITALPIDCONTROLLERUSUALLYCANBEEXPRESSEDASUKKPEKKIΣEIKDECK（4）INFUZZYREASONING,THEINPUTVARIABLESAREEANDEC,ANDTHEOUTPUTVARIABLESAREΔKP,ΔKI,ΔKDTHEPIDCONTROLPARAMETERSAREGIVENASKPKP0ΔKP，KIKI0ΔKI，KDKD0ΔKD（5）WHERE,KP0,KI0,KD0ARETHEINITIALSETTINGVALUESOFKD,KI,KDPIDCONTROLPARAMETERSBTHERULETABLESOFFUZZYCONTROLFROMTHECHARACTERISTICSOFPIDCONTROLTOKNOW,THESTRONGINTEGRALACTIONILEADSTOBIGOVERSHOOT,FASTRESPONSETHESTRONGDERIVATIVEACTIONDHASGOODSTABILITY,SMALLOVERSHOOT,SMALLABILITYOFANTIINTERFERENCEACCORDINGTOEANDECATDIFFERENTSTAGES,THESETTINGPRINCIPLEOFPIDCONTROLPARAMETERSAREGIVEN11,181WHENTHECONTROLLEDVARIABLESARECLOSETOSETTINGVALUES,THEINTEGRALACTIONWITHTHESAMESIGNASECCANAVOIDSOVERSHOOTANDOSCILLATION,ANDISBENEFICIALTOTHECONTROLWHENTHECONTROLLEDVARIABLESAREFARFROMSETTINGVALUES,THEINTEGRALACTIONWITHTHEOPPOSITESIGNASECANREDUCEOVERSHOOTANDAVOIDOSCILLATION2ATINITIALSTAGEOFPIDPARAMETERADJUSTMENT,ITCANAVOIDOVERSHOOTANDINCREASERESPONSESPEEDTHATKPISPROPERLYLARGEANDKIISSMALLORZEROATMIDDLESTAGE,LETTHEVALUESOFKPANDKIBEMODERATEWHILETAKINGACCOUNTTOSTABILITYANDCONTROLPRECISIONATLASTSTAGE,ITCANELIMINATETHEERRORSANDREDUCEOVERSHOOTTHATKPISREDUCEDANDKIISINCREASEDPROPERLY3DIFFERENTIALCOEFFICIENTKDCANINHIBITCHANGEOFCONTROLLEDVARIABLE,SHORTENSTEADYTIME,REDUCESTEADYSTATEERRORITISASUPPLEMENTTOKP,KITHEFUZZYSUBSETSOFINPUTANDOUTPUTVARIABLESOFFUZZYCONTROLLERARERESPECTIVELYE,EC,ΔKP,ΔKI,ΔKDTHELANGUAGEVALUESOFTHESEFUZZYVARIABLESAREIN{NB,NM,NS,ZO,PS,PM,PB},ANDTHEMEMBERSHIPFUNCTIONSOFTHESEFUZZYVARIABLESAREALLSENSITIVELYTRIGONOMETRICFUNCTIONSWITHTHEVALUEFIELDIN3,3BYFUZZYREASONINGANDTESTMODIFICATIONBASEDONTHEABOVESETTINGPRINCIPLE,THEFUZZYCONTROLRULESOFTHESELFTUNINGPARAMETERFUZZYPIDCONTROLSYSTEMAREOBTAINEDANDSHOWNINTABLEIIITALLEIIFUZZYCONTROLRULESOFBP0CONTROLLOOPFIG5THESTRUCTUREOFSELFTUNINGFUZZYPIDCONTROLLER

簡介：PAPERACCEPTEDFORPRESENTATIONAT2003IEEEBOLOGNAPOWERTECHCONFERENCE,JUNE2326,BOLOGNA,ITALYDETECTIONOFTHEINTERMITTENTEARTHFAULTSINCOMPENSATEDMVNETWORKJOZEFLORENC,KAZIMIERZMUSIEROWICZ“,ANDRZEJKWAPISZ“ABSTRACTTHEEXPERIENCEACQUIREDFROMTHEPOLISHMEDIUMVOLTAGEPOWERDISTRIBUTIONNETWORKSSHOWSTHENNRELIABILITYOFTHELOCALIZATIONCRITERIONSAPPLIEDTOTHEINTERMITTENTEARTHFAULTSITRESULTSFMMTHELACKOFSTABILITYANDLOWPOWERLEVELOFTHEMEASURINGSIGNALSFALLINGOFTENDOWNBELOWTHEPROTECTIONSSTARTUPLEVELINTHEPAPER,ANEWADAPTIVEALGORITHMBASEDONTHEWAVELETANALYSISENABLINGDETECTIONOFSPECIFICDYNAMICSOFTHEMEASURINGSIGNALDURINGINTERMITTENTEARTHFAULTSISPRESENTEDTHEALGORITHMWASANALYZEDUTILIZINGTHESIGNALSGENERATEDINTHEEMTPPROGRAMPACKAGEIINTRODUCTIONINGENERAL,THEMVDISTRIBUTIONNETWORKSINPOLANDOPERATEWITHTHENEUTRALPOINTGROUNDEDTHROUGHTHECOILTOCOMPENSATETHECAPACITIVESHORTCIRCUITCURRENTTOTHEEARTHITREFERSMAINLYTOTHERURALAREANETWORKSWHERETHELINESARETHEOVERHEADONESSUCHNETWORKSARECHARACTERIZEDBYLARGENUMBEROFTHEEARTHFAULTSEXCEEDING90OFALLRECORDEDFAULTSDUETOTHERELATIVELYHIGHCROSSRESISTANCEATTHEDEFECTSLOCATIONRFASWELLASTOTHEEFFECTSOFTHEWEATHERPHENOMENASUCHASDISCHARGES,GUSTSOFWIND,HIGHANDLOWTEMPERATURESRESULTINGINTHERUPTUREOFTHELINECONDUCTORSCONTINUITY,THEEARTHFAULTSOCCURCHARACTERISTICSOFTHESEFAULTSMAKESIMPOSSIBLETHEDETECTIONANDLOCALIZATIONOFSUCHDISTURBANCE3THEFOLLOWINGFAULTTYPESCANBEENCOUNTEREDTOTHEDISCUSSEDFAULTSGROUPANACTUALFAULTCANSHOWEITHERONEORALLOFTHELISTEDFEATURESINTHEPAPER,THEANALYSISISLIMITEDTOTHEAUTOMATICPROTECTIVEUNITSOPERATIONDURINGINTERMITTENTFAULTSTOASSESSTHEPROTECTIONSOPERABILITY,THELEVELSANDFEATURESOFMEASURINGSIGNALSWHICHCANOCCURDURINGTHEFAULTARETOBEIDENTIFIEDTHEMOSTIMPORTANTSIGNALINDICATINGOCCURRENCEOFTHEINTERMITTENTEARTHFAULTINTHENETWORKISAZEROVOLTAGECOMPONENTTHEVALUESOFWHICHISOFTENFOUNDBYADDINGTHEINSTANTVALUESOFPHASEVOLTAGESTHECRITERIONVALUEOFTHEFAULTLOCALIZATIONCANHEZEROCURRENTCOMPONENT,BRESISTANCEFAULTSOFHIGHCROSSRESISTANCE,RF,BREAKINTHELIVEWIRESHORTCIRCUITONTHERECEIVERSIDE,FAULTSBEINGBROKENCYCLICALLYANDNONCYCLICALLYPOWEROFTHEZEROCURRENTCOMPONENT,B,ANDZEROVOLTAGECOMPONENT,UO,PHASESHIFTANGLEBETWEENTHEZEROCURRENTANDVOLTAGECOMPONENTS,ZEROADMITTANCECOMPONENT,YO,ORITSCOMPONENTSACTIVEGOORREACTIVEBOHOWEVER,THECRITERIONVALUESASLISTEDABOVEAREOFTENUNRELIABLEWHENTHEINTERMITTENTEARTHFAULTOCCURS11MODELOFNETWORKFORMODELINGANDSTUDIESOFTHEEARTHFAULTPHENOMENAACCOMPANYINGTHEINTERMITTENTEARTHFAULTS,ATYPICALMEDIUMVOLTAGEBALANCEDNETWORKHASBEENCHOSENTHESCHEMEOFMODELEDNETWORKISSHOWNINFIG1IFLG1MEDIUMVOLTAGENETWORKSCHEMETHEFAULTSWEREMODELEDANDSIMULATEDUSINGTHEEMTP/ATPPROGRAMPACKAGECHOSENPARAMETERSOFNETWORKASSUMEDFORSIMULATIONPURPOSESARESHOWNINTABLE1RMMIMODEUED15KVNETWORKPARAMERRRSNETWORKCAPACITYCURRENTBFAULTLINECAPACITYCNNENTLA1101,3AI10,6ADECOMPENSATIONLEVELIISCMSSRESISTANCERPINNTHEWORKHASBEENSUPPORTEDBYTHESTATECOMMITTEEFORSCIENTIFICRESEARCHDS41612ANDFAMEWORKPROGRAMME5FP5THEAUTHORSAREFROMPOZNANUNIVERSITYOFRECHNOIAGYJO2EFLORENCPUTPOZNANPLKAZIMIEMUSIEMWICIPNLPOZNANPLANDRZEJKWAPISRAJAXEPEPUTPZNANPLINTHEMODELTHEASSUMPTIONWASMADETHATTHEFAULTSOCCURINALINEWITHTOGROUNDCAPACITIVECURRENTOF10,6AANDAMODERATEDPOWERLOADOF150KWTHEFOLLOWINGFAULTTYPESHAVEBEENCONSIDERED0780379675/03/170002003IEEEIIIITIMEMRAG1HTNENTEARTHFAULTDTYPEINLINE1ADISCRIMINATIONOFTHESHORTCIRCUITEDLINEINCASEWHENTHEPAUSEBETWEENSUCCESSIVEFAULTSISRELATIVELYLONGLONGTIMET,CANDDTYPESCANHEMOREDIFFICULTTHERUNSRELATEDTOSUCHFAULTSARESHOWNINFIGURES6AND7DURINGENTIREDURATIONOFFAULT,THEVOLTAGEZEROCOMPONENTTOWHICHTHESTARTUPSIGNALS,ISPROPORTIONALREMAINSATTHEHIGHLEVELITMEANSTHATTHEFAULTDETECTIONSHOULDNOTBEDIFFICULTDUETOTHERELATIVELYLOWATTENUATIONOFTHEVOLTAGETRANSIENTSAFTERINSTANTANEOUSDISAPPEARINGOFTHEFAULTHOWEVER,THEPROBLEMSCOULDARISEWITHDAMAGEDLINEDISCRIMINATIONDUETOTHEFEATURESOFTHETRANSIENTPROCESSINTHENETWORKRESULTINGFROMTHECYCLICARCIGNITIONSINTHEFAULTLOCATION,THEDAMAGEDLINEADMITTANCEFALLSCYCLICALLYDOWNTOTHEUNDAMAGEDLINEADMITTANCELEVELITREFERSTOBOTHTHECTYPEANDDTYPEFAULTSINSUCHACASE,ANIMPROPEROPERATIONOFTHEADMITTANCECRITERIONRELATEDPROTECTIONCANBEEXPECTEDBETTEROPPORTUNITIESOPENWHENUSINGTHEWAVELETEXPANSIONS2,4,151,ESPECIALLYTHEMULTIRESOLUTIONDECOMPOSITION6OFTHEMEASURINGSIGNALSWDIVMULTIRFSOLUTIONWAVELETANALYSISAMAINTOOLOFTHEWAVELETANALYSISINTHEPROPOSEDAPPLICATIONISTHEMULTIRESOLUTIONDECOMPOSITIONOFMEASURINGSIGNALSREALIZEDBYTHEMULTISTAGESETOFTHEWAVELETCOMPLEMENTARYFILTERSHIGHPASSWAVELETSANDLOWPASSSCALINGFUNCTIONSTHECALCULATINGPROCEDURELEADINGTOTHEDECOMPOSITIONISCALLEDTHEMALLUTALGORITHMLTHEITERATIONPROCESSOFCREATINGTHEMULTIRESOLUTIONSIGNALREPRESENTATIONCANHEPRESENTEDINTHEFORMOFTHEWAVELETSIGNALDECOMPOSITIONTREEASSHOWNINFIG8ATANYITERATIVESTEP,THEANALYSEDSIGNALISFILTEREDTHENUMBEROFITERATIVESTEPSISUNLIMITEDINFIG8,NSTEPSHAVEBEENASSUMEDEACHITERATIONRESULTSINBOTHTHEHIGHFREQUENCYCOMPONENTCALLEDUDETAILDIWHICHISNOMOREFILTEREDDURINGSUCCESSIVEITERATIVESTEPS,ANDTHELOWFREQUENCYCOMPONENTAIOFANALYSEDORIGINALSIGNALS,CALLEDANAPPROXIMAFIONTHUS,THESIGNALDECOMPOSITIONPROCESSHASAFORMOFTHEMULTILEVELITERATIVEPROCESSCARRIEDOUTONTHELOWPASFILTRATIONCHANNEL,ANDTHESUCCESSIVEAPPROXIMATIONSARESUBJECTTOTHESUCCESSIVEDECOMPOSITIONWHENCHOOSINGTHEMOTHERWAVELETFORANALYSISOFMEASURINGSIGNALSSHOWNINFIGURES5,6AND7,THEKNOWNRULEHASBEENTAKENINTOACCOUNTTHESMOOTHSHAPEWAVELETSTHEMORLETSWAVELET,FOREXAMPLEAREOFBETTERRESOLUTIONWHENANALYSINGTHESIGNALFREQUENCYSPECTRUM,IETHEYHAVEBETTERLOCALIZATIONOFFREQUENCYCOMPONENTSALONGTHEFREQUENCYAXIS,WHILETHEDISCONTINUOUSLYSHAPEDWAVELETSTHEHAARSWAVELET,FOREXAMPLEHAVEBETTERRESOLUTIONALONGTHETIMEAXISREFERRINGTOTHEOVERVIEWOFPROPERTIESOFMANYWAVELETSTYPES,THEAUTHORSDREWACONCLUSIONTHATTHEHAARSWAVELETSPROPERTIESMEETINTHEHESTWAYTHEREQUIREMENTSOFTHECONSIDEREDAPPLICATION,REGARDINGBOTHTHEMETROLOGICALASPECTSASTHESPEEDOFREALTIMECALCULATIONSCARRIEDOUTINPROTECTIONSTHEBASICHAARSWAVELETISDEFINEDASFOLLOWS1FOROFIF/1FORTI710FOROTHERFANDGENERATESASETOFWAVELETSWITHELEMENTSAS1F/,,T2““2“TNFORM,N,2,L,O,1,2,8ADVANTAGEOFTHEVMFUNCTIONINPROPOSEDAPPLICATIONISTHEIRGOODLOCALIZATIONASFORANINFINITELYPRECISELOCALIZATIONINTIMEISOBTAINEDENABLINGARBITRARYACCURACYOFLOCALIZATIONOFTHEFUNCTIONDISCONTINUITYESPECIALLYTHAT

簡介：ASPECIALPROTECTIONSCHEMEFORVOLTAGESTABILITYPREVENTIONTARAALZAHAWISTUDENTMEMBER,IEEEMOHINDARSSACHDEVLIFEFELLOW,IEEEGRAMAKRISHNAMEMBER,IEEEPOWERSYSTEMRESEARCHGROUPUNIVERSITYOFSASKATCHEWANSASKATOON,SKS7N5A9,CANADAABSTRACTVOLTAGEINSTABILITYISCLOSELYRELATEDTOTHEMAXIMUMLOADABILITYOFATRANSMISSIONNETWORKTHEENERGYFLOWSONTHETRANSMISSIONSYSTEMDEPENDONTHENETWORKTOPOLOGY,GENERATIONANDLOADS,ANDONTHEAVAILABILITYOFSOURCESTHATCANGENERATEREACTIVEPOWERONEOFTHEMETHODSUSEDFORTHISPURPOSEISTHEVOLTAGEINSTABILITYPREDICTORVIPTHISRELAYMEASURESVOLTAGESATASUBSTATIONBUSANDCURRENTSINTHECIRCUITCONNECTEDTOTHEBUSFROMTHESEMEASUREMENTS,ITESTIMATESTHETHéVENIN’SEQUIVALENTOFTHENETWORKFEEDINGTHESUBSTATIONANDTHEIMPEDANCEOFTHELOADBEINGSUPPLIEDFROMTHESUBSTATIONTHISPAPERDESCRIBESANEXTENSIONTOTHEVIPTECHNIQUEINWHICHMEASUREMENTSFROMADJOININGSYSTEMBUSESANDANTICIPATEDCHANGEOFLOADARETAKENINTOCONSIDERATIONASWELLKEYWORDSMAXIMUMLOADABILITYVOLTAGEINSTABILITYVIPALGORITHM1INTRODUCTIONDEREGULATIONHASFORCEDELECTRICUTILITIESTOMAKEBETTERUSEOFTHEAVAILABLETRANSMISSIONFACILITIESOFTHEIRPOWERSYSTEMTHISHASRESULTEDININCREASEDPOWERTRANSFERS,REDUCEDTRANSMISSIONMARGINSANDDIMINISHEDVOLTAGESECURITYMARGINSTOOPERATEAPOWERSYSTEMWITHANADEQUATESECURITYMARGIN,ITISESSENTIALTOESTIMATETHEMAXIMUMPERMISSIBLELOADINGOFTHESYSTEMUSINGINFORMATIONABOUTTHECURRENTOPERATIONPOINTTHEMAXIMUMLOADINGOFASYSTEMISNOTAFIXEDQUANTITYBUTDEPENDSONVARIOUSFACTORS,SUCHASNETWORKTOPOLOGY,AVAILABILITYOFREACTIVEPOWERRESERVESANDTHEIRLOCATIONETCDETERMININGTHEMAXIMUMPERMISSIBLELOADING,WITHINTHEVOLTAGESTABILITYLIMIT,HASBECOMEAVERYIMPORTANTISSUEINPOWERSYSTEMOPERATIONANDPLANNINGSTUDIESTHECONVENTIONALPVORVQCURVESAREUSUALLYUSEDASATOOLFORASSESSINGVOLTAGESTABILITYANDHENCEFORFINDINGTHEMAXIMUMLOADINGATTHEVERGEOFVOLTAGECOLLAPSE1THESECURVESAREGENERATEDBYRUNNINGALARGENUMBEROFLOADFLOWCASESUSING,CONVENTIONALMETHODSWHILESUCHPROCEDURESCANBEAUTOMATED,THEYARETIMECONSUMINGANDDONOTREADILYPROVIDEINFORMATIONUSEFULINGAININGINSIGHTINTOTHECAUSEOFSTABILITYPROBLEMS2TOOVERCOMETHEABOVEDISADVANTAGESSEVERALTECHNIQUESHAVEBEENPROPOSEDINTHELITERATURE,SUCHASBIFURICATIONTHEORY3,ENERGYMETHOD4,EIGENVALUEMETHOD5,MULTIPLELOADFLOWSOLUTIONSMETHOD6ETCREFERENCE7PROPOSEDASIMPLEMETHOD,WHICHDOESNOTREQUIREOFFLINESIMULATIONANDTRAININGTHEVOLTAGEINDICATORPREDICTORVIPMETHODIN7ISBASEDONLOCALMEASUREMENTSVOLTAGEANDCURRENTANDPRODUCESANESTIMATEOFTHESTRENGTH/WEAKNESSOFTHETRANSMISSIONSYSTEMCONNECTEDTOTHEBUS,ANDCOMPARESITWITHTHELOCALDEMANDTHECLOSERTHELOCALDEMANDISTOTHEESTIMATEDTRANSMISSIONCAPACITY,THEMOREIMMINENTISTHEVOLTAGEINSTABILITYTHEMAINDISADVANTAGEOFTHISMETHODISINTHEESTIMATIONOFTHETHéVENIN’SEQUIVALENT,WHICHISOBTAINEDFROMTWOMEASUREMENTSATDIFFERENTTIMESFORAMOREEXACTESTIMATION,ONEREQUIRESTWODIFFERENTLOADMEASUREMENTSTHISPAPERPROPOSESANALGORITHMTOIMPROVETHEROBUSTNESSOFTHEVIPALGORITHMBYINCLUDINGADDITIONALMEASUREMENTSFROMSURROUNDINGLOADBUSESANDALSOTAKINGINTOCONSIDERATIONLOCALLOADCHANGESATNEIGHBORINGBUSES2PROPOSEDMETHODOLOGYTHEVIPALGORITHMPROPOSEDINTHISPAPERUSESVOLTAGEANDCURRENTMEASUREMENTSONTHELOADBUSESANDASSUMESTHATTHEIMPEDANCEOFINTERCONNECTINGLINESZ12,Z13AREKNOWN,ASSHOWNINFIGURE1THECURRENTFLOWINGFROMTHEGENERATORBUSTOTHELOADBUSISUSEDTOESTIMATETHéVENIN’SEQUIVALENTFORTHESYSTEMINTHATDIRECTIONSIMILARLYTHECURRENTFLOWINGFROMOTHERLOADBUSFIGURE2ISUSEDTOESTIMATETHéVENIN’SEQUIVALENTFROMOTHERDIRECTIONTHISRESULTSINFOLLOWINGEQUATIONSFIGURE3NOTETHATTHECURRENTCOMINGFROMTHESECONDLOADBUSOVERTHETRANSMISSIONLINEWASKEPTOUTOFESTIMATIONINORIGINALVIPALGORITHM111112211211111??????THZTHEZVZTHZLZV1122112111212122??????THZTHEZVZTHZLZV21111111EITHZVTHZTHE???32122122EITHZVTHZTHE???40780388860/05/2000?2005IEEECCECE/CCGEI,SASKATOON,MAY2005545THZLOADVTHEI8VANDIAREDIRECTLYAVAILABLEFROMMEASUREMENTSATTHELOCALBUSEQUATION8CANBEEXPRESSEDINTHEMATRIXFORMASSHOWNBELOW??????????00IVRV????????????THXTHRITHERTHE????????????000010000001RIIIIIRI9BAX10THEUNKNOWNPARAMETERSCANBEESTIMATEDFROMTHEFOLLOWINGEQUATIONBTAAXTA11NOTETHATALLOFTHEABOVEQUANTITIESAREFUNCTIONSOFTIMEANDARECALCULATEDONASLIDINGWINDOWOFDISCRETEDATASAMPLESOFFINITE,PREFERABLYSHORTLENGTHTHEREAREADDITIONALREQUIREMENTSTOMAKETHEESTIMATIONFEASIBLETHEREMUSTBEASIGNIFICANTCHANGEINLOADIMPEDANCEINTHEDATAWINDOWOFATLEASTTWOSETOFMEASUREMENTSFORSMALLCHANGESINTHéVENIN’SPARAMETERSWITHINAPARTICULARDATAWINDOW,THEALGORITHMCANESTIMATEPROPERLYBUTIFASUDDENLARGECHANGEOCCURSTHENTHEPROCESSOFESTIMATIONISPOSTPONEDUNTILTHENEXTDATAWINDOWCOMESINTHEMONITORINGDEVICEBASEDONTHEABOVEPRINCIPLECANBEUSEDTOIMPOSEALIMITONTHELOADINGATEACHBUS,ANDSHEDSLOADWHENTHELIMITISEXCEEDEDITCANALSOBEUSEDTOENHANCEEXISTINGVOLTAGECONTROLLERSCOORDINATEDCONTROLCANALSOBEOBTAINEDIFCOMMUNICATIONISAVAILABLEONCEWEHAVETHETIMESEQUENCEOFVOLTAGEANDCURRENTWECANESTIMATEUNKNOWNSBYUSINGPARAMETERESTIMATIONALGORITHMS,SUCHASKALMANFILTERINGAPPROACHDESCRIBED622VOLTAGESTABILITYMARGINSANDTHEMAXIMUMPERMISSIBLELOADINGSYSTEMREACHESTHEMAXIMUMLOADPOINTWHENTHECONDITION|ZLOAD||ZTHEV|ISSATISFIEDFIGURE5THEREFORETHEVOLTAGESTABILITYBOUNDARYCANBEDEFINEDBYACIRCLEWITHARADIUSOFTHETHéVENIN’SIMPEDANCEFORNORMALOPERATIONTHE|ZTHEV|ISSMALLERTHAN|ZLOAD|IEITISOUTSIDETHECIRCLEANDTHESYSTEMOPERATESONTHEUPPERPARTORTHESTABLEREGIONOFACONVENTIONALPVCURVE2HOWEVER,WHEN|ZTHEV|EXCEEDS|ZLOAD|THESYSTEMOPERATESONTHELOWERPARTORUNSTABLEREGIONOFTHEPVCURVE,INDICATINGTHATVOLTAGECOLLAPSEHASALREADYOCCURREDATTHEMAXIMUMPOWERPOINT,THELOADIMPEDANCEBECOMESSAMEASTHETHéVENIN’SZLZTHEVTHEREFORE,FORAGIVENLOADIMPEDANCEZLOAD,THEDIFFERENCEBETWEENZTHEVANDZLOADCANBECONSIDEREDASASAFETYMARGINHENCETHEVOLTAGESTABILITYMARGINVSMDUETOIMPEDANCESCANBEEXPRESSEDASVSMZWHERESUBSCRIPTZDENOTESTHEIMPEDANCETHEREFOREWEHAVELOADZTHEVZLOADZZVSM?12THEABOVEEQUATIONASSUMESTHATBOTHLOADIMPEDANCESZ1,Z2AREDECREASINGATASTEADYRATE,SOTHEPOWERDELIVEREDTOBUS1WILLINCREASEACCORDINGTOEQUATION7HOWEVERONCEITREACHESTHEPOINTOFCOLLAPSEPOWERSTARTSTODECREASEAGAINNOWASSUMETHATBOTHLOADSAREFUNCTIONSOFTIMETHEMAXIMUMCRITICALLOADINGPOINTISTHENGIVENBYEQUATION130DT1DSCRITICAL1S13EXPRESSINGVOLTAGESTABILITYMARGINDUETOLOADAPPARENTPOWERASSVSM,WEHAVECRITICALSLOADSCRITICALSSVSM?14NOTETHATBOTHVSMZANDVSMSARENORMALIZEDQUANTITIESANDTHEIRVALUESDECREASEASTHELOADINCREASESATTHEVOLTAGECOLLAPSEPOINT,BOTHTHEMARGINSREDUCETOZEROANDTHECORRESPONDINGLOADISCONSIDEREDASTHEMAXIMUMPERMISSIBLELOADINGFIG5VIPALGORITHM23ADVANTAGESOFTHEPROPOSEDVIPALGORITHMBYINCORPORATINGTHEMEASUREMENTSFROMOTHERLOADBUSESFIGURE3,THEPROPOSEDVIPALGORITHMACHIEVESAMOREACCURATEVALUEOFZTHEVTHEONLINETRACKINGOFZTHEVISUSEDTOTRACKSYSTEMCHANGESTHEPROPOSEDIMPROVEMENTSINTHEVIPALGORITHMWILLRESULTINBETTERCONTROLACTIONFORPOWERSYSTEMVOLTAGESTABILITYENHANCEMENTTHECONTROLMEASURESARENORMALLYSHUNTREACTORDISCONNECTION,SHUNTCAPACITORCONNECTION,SHUNTVARCOMPENSATIONBYMEANSOFSVC’SANDSYNCHROUNSCONDENSERS,STARTINGOFGASTURBINES,LOWPRIORITYLOADDISCONNECTION,ANDSHEDDINGOFLOWPRIORITYLOAD8FIGURE6SHOWSTHEMOSTCOMMONLYUSEDREMEDIALACTIONSTHEVZLOAZVSM547

簡介：OPTICALFIBERTECHNOLOGY142008149–153WWWELSEVIERCOM/LOCATE/YOFTEEFFECTOFCABLINGONPOLARIZATIONMODEDISPERSIONINOPTICALFIBERRIBBONCABLESKUNIHIROTOGE?,KAZUOHOGARINTTACCESSSERVICESYSTEMLABORATORIES,NTTCORPORATION,171,HANABATAKE,TSUKUBACITY,IBARAKI3050805,JAPANRECEIVED5APRIL2007REVISED27JULY2007AVAILABLEONLINE3DECEMBER2007ABSTRACTTHISPAPERTHEORETICALLYANDEXPERIMENTALLYINVESTIGATESANEFFECTOFCABLINGONPOLARIZATIONMODEDISPERSIONPMDINRIBBONFIBERSHELICALLYSTRANDEDINOPTICALFIBERRIBBONCABLESAIMEDATDESIGNINGLOWPMDRIBBONCABLESBASEDONTHEBIREFRINGENCEMODELFOCUSEDONTHECHANGEINTHEBIREFRINGENCEWHENRIBBONFIBERSARECABLED,THEHELICALPITCHOFOPTICALFIBERRIBBONCABLESISDESIGNEDTOMINIMIZETHEMAXIMUMPMDINTHECABLESALOWPMDCHARACTERISTICISCONFIRMEDINOPTICALFIBERRIBBONCABLEWITHAPPROXIMATELYTHEOPTIMALHELICALPITCH?2007ELSEVIERINCALLRIGHTSRESERVEDKEYWORDSOPTICALFIBERCABLEOPTICALFIBERRIBBONPOLARIZATIONMODEDISPERSION1INTRODUCTIONTHEDEMANDFORGREATERTRANSMISSIONCAPACITYISGROWINGRAPIDLYASARESULTOFTHEINCREASEINTHENUMBEROFBROADBANDSERVICESPROVIDEDBYTHEINTERNETANDTHEBITRATEHASBEENINCREASINGTOMEETTHISDEMANDASTHEBITRATEHASINCREASED,POLARIZATIONMODEDISPERSIONPMDHASBECOMEAMAJORFACTORLIMITINGTHETRANSMISSIONLENGTHANDHASATTRACTEDINCREASINGATTENTION1OPTICALFIBERRIBBONCABLESAREWIDELYUSEDINBOTHACCESSANDTRUNKNETWORKSBECAUSEOFTHEIRHIGHCOUNTCOMPACTNESS,EASEOFFIBERIDENTIFICATIONANDCAPACITYFORMASSSPLICINGSEVERALSTUDIESHAVEINVESTIGATEDTHEPMDINOPTICALFIBERRIBBONSANDCABLES2–5ITHASBEENREPORTEDTHATTHEINNERPAIROFFIBERSIN4FIBERRIBBONSHASAHIGHPMDAFINITEELEMENTANALYSISHASALSOBEENPERFORMEDTOMODELTHESTRESSDISTRIBUTIONINOPTICALFIBERRIBBONS,ANDINDUCEDBIREFRINGENCEWASFOUNDTOBEHIGHFORTHEINNERFIBERSINTHERIBBONSASARESULTOFTHERIBBONCOATING2SUCHAHIGHPMDHASALSOBEENFOUNDINCABLESHOWEVER,THISHIGHPMDHASOFTENAPPEAREDINTHEOUTERFIBERSINRIBBONS3,4THESERESULTSLEADUSTOBELIEVETHATTHECHANGEINTHEBIREFRINGENCEINDUCEDBYCABLINGWOULDSIGNIFICORRESPONDINGAUTHORFAX81298526142EMAILADDRESSTOGEANSLNTTCOJPKTOGECANTLYAFFECTTHEPMD,ANDWOULDDEPENDSTRONGLYONTHECABLESTRUCTUREHOWEVER,THECHANGEINTHEBIREFRINGENCEINDUCEDBYCABLINGHASREMAINEDUNCLEARINORDERTOACHIEVELOWPMDCABLES,ITISIMPORTANTTODESIGNTHESTRUCTUREOFOPTICALFIBERRIBBONCABLESTAKINGTHEEFFECTOFCABLINGONTHEBIREFRINGENCEINOPTICALFIBERRIBBONSINTOCONSIDERATIONINTHISPAPER,WEINVESTIGATETHEEFFECTOFCABLINGONPMDINOPTICALFIBERRIBBONCABLESWITHHELICALLYSTRANDEDRIBBONSTHEORETICALLYANDEXPERIMENTALLYFIRST,WECOMPARENUMERICALLYCALCULATEDANDMEASUREDRESULTSTOCONFIRMTHEMODELINTHENUMERICALCALCULATION,WEUSEABIREFRINGENCEMODELFOCUSEDONTHECHANGEINTHEBIREFRINGENCEINRIBBONFIBERSINDUCEDBYTENSION,BENDINGANDTWISTINGWHENRIBBONFIBERSARECABLEDTHEN,WEDISCUSSTHEEFFECTOFCABLESTRUCTUREONTHEPMDINOPTICALFIBERRIBBONCABLES,ANDPOINTOUTTHATTHEHELICALPITCHOFCABLESCANBEOPTIMIZEDTOMINIMIZETHEMAXIMUMPMDINOPTICALFIBERRIBBONCABLESWEALSOMEASURETHEPMDINOPTICALFIBERRIBBONCABLESWITHAPPROXIMATELYTHEOPTIMALHELICALPITCHTOREVEALTHEVALIDITYOFTHEDESIGNANDREPORTTHERESULT2THEORETICALBACKGROUND21BIREFRINGENCECAUSEDBYCABLINGWECONSIDEREDTWOTYPESOFOPTICALFIBERRIBBONCABLECOMPOSEDOF4FIBERRIBBONSASSHOWNINFIG1ONEWAS100FIBER10685200/–SEEFRONTMATTER?2007ELSEVIERINCALLRIGHTSRESERVEDDOI101016/JYOFTE200709010KTOGE,KHOGARI/OPTICALFIBERTECHNOLOGY142008149–153151TABLE1DIMENSIONSOFOPTICALFIBERRIBBONSANDCABLESUSEDINTHEEXPERIMENTSANDCALCULATIONSCABLEACABLEBFIBERRIBBONCOUNT100254010PITCHRADIUSA25MM0MMHELICALPITCHP500MM350MMFIBERDIAMETER2R125ΜM125ΜMRIBBONWIDTH11MM11MMRIBBONTHICKNESS03MM03MMFIG2POLARIZATIONSENSITIVEOPTICALTIMEDOMAINREFLECTOMETRYSETUPUSEDTOMEASUREBEATLENGTHTABLE1SUMMARIZESTHETYPICALDIMENSIONSOFTHERIBBONSANDTHECABLESWEUSED32PMDANDBEATLENGTHMEASUREMENTTHEPMDWASMEASUREDFORRIBBONFIBERSANDCABLEDRIBBONFIBERSBYTHEJONESMATRIXEIGENANALYSISMETHODINTHE1520–1630NMWAVELENGTHRANGETHEBEATLENGTHLBWASMEASUREDWITHAPOLARIZATIONSENSITIVEOPTICALTIMEDOMAINREFLECTOMETRYPOTDRTECHNIQUE13,14FIGURE2SHOWSTHEMEASUREMENTSETUPADISTRIBUTEDFEEDBACKLASERDIODEDFBLDWITHANARROWLINEWIDTHOF160MHZWASUSEDTOAVOIDDEPOLARIZATIONINWAVELENGTH15WEOPERATEDTHEDFBLDAT1550NM,ANDVARIEDTHEWAVELENGTHONARANGEOF1NMDURINGAVERAGINGINORDERTOREDUCETHECOHERENCENOISE13THEDFBLDWASEXTERNALLYMODULATEDBYALINBO3LNMODULATORTHEPULSEWIDTHWASSETAT10NS,WHICHCORRESPONDSTOASPATIALRESOLUTIONOF1MTOOBTAINANOPTICALPULSEWITHAHIGHPEAKPOWER,WEUSEDANOPTICALAMPLIFIERANDELIMINATEDAMPLIFIEDNOISEBYUSINGANACOUSTOOPTICMODULATORALINEARPOLARIZERWASUSEDINFRONTOFTHEFIBERINPUTENDTHERAYLEIGHBACKSCATTERINGLIGHTTHATPASSEDTHROUGHTHEPOLARIZERWASDETECTEDANDTHENAVERAGEDWEANALYZEDITSPOWERFLUCTUATIONTOOBTAINTHEBEATLENGTHTHEBEATLENGTHWASCALCULATEDFROMTHEPEAKFOURIERFREQUENCYOFTHEPOWERSPECTRUMOFTHEFLUCTUATIONASDESCRIBEDIN14THECOUPLINGLENGTHLCINCABLESCANBEESTIMATEDFROMTHEMEASUREDPMDANDBEATLENGTHINCABLESTHESTRAINDIFFERENCEBETWEENRIBBONFIBERSANDCABLEDRIBBONFIBERS?ΕWASMEASUREDBYUSINGABRILLOUINOTDR164RESULTSANDDISCUSSION41COMPARISONOFCALCULATEDANDMEASUREDPMDFIGURE3COMPARESTHEPMDREDUCTIONFACTORSPMDRFSOBTAINEDBYMEASUREMENTANDCALCULATION,WHERETHEPMDRFSFIG3COMPARISONOFCALCULATEDANDMEASUREDPMDREDUCTIONFACTORSPMDRFSFORCABLESAANDBTHEOPENANDCLOSEDSYMBOLSSHOWTHEPMDRFSFORTHEOUTERANDINNERFIBERSINRIBBONS,RESPECTIVELYTABLE2MEASUREDBEATLENGTHSANDPMDINOPTICALFIBERRIBBONSBEATLENGTHLBMPMDINRIBBONSPS/RKMOUTERFIBERSMAX26016MEAN17010MIN2005INNERFIBERSMAX17035MEAN6023MIN2011NOTENUMBEROFMEASUREDFIBERSIS10FOREACHFIG4CALCULATEDPMDASAFUNCTIONOFTHEHELICALPITCHANDLATERALSTRESSAREDEFINEDBYTHEPMDINCABLEDRIBBONFIBERSWHICHISNORMALIZEDBYTHEMEASUREDPMDINRIBBONFIBERSSINCETHEPMDINTHELONGLENGTHREGIMEEXHIBITSAMAXWELLDISTRIBUTIONOWINGTORANDOMMODECOUPLING,WEDEFINEDTHECALCULATEDPMDASTHEMEANPMDOBTAINEDFROMTHECALCULATIONOF100INSTANCESTHEBEATLENGTHINRIBBONFIBERSANDTHESTRAINDIFFERENCEWEUSEDFORTHECALCULATIONAREBASEDONMEASUREDRESULTSWEFOUNDTHATTHECALCULATEDANDMEASUREDRESULTSAGREEDWELL,WHICHINDICATESTHATTHEBIREFRINGENCECHANGECAUSEDBYCABLINGCANBEMODELEDASDESCRIBEDINTHEPREVIOUSSECTION42EFFECTSOFCABLINGONPMDINOPTICALFIBERRIBBONSTABLE2SHOWSTHEMEASUREDBEATLENGTHSANDPMDINRIBBONFIBERSITCANBESEENTHATTHEMEANBEATLENGTHINTHEINNERFIBERSINTHERIBBONSISABOUTTHREETIMESSHORTERTHANTHOSEINTHEOUTERFIBERS,WHICHAGREESWITHTHERESULTSREPORTEDIN2FIG

簡介：AUTOMATICIDENTIFICATIONOFGEOMETRICCONSTRAINTSINMECHANICALASSEMBLIESSHMULLINS?ANDDCANDERSON?MECHANICALASSEMBLIESOFMANUFACTUREDCOMPONENTSINVOLVESETSOFRELATIONSBETWEENMATINGSURFACESANDFUNCTIONALCHARACTERISTICSCOMPONENTSMUSTFITTOGETHERTOASSEMBLEANDFUNCTIONPROPERLY,PLACINGCONSTRAINTSONTHEALLOWABLEVALUESOFTHECOMPONENTDIMENSIONSKINEMATICMOTIONOFTHECOMPONENTSISOFTENNECESSARY,RESULTINGINOTHERGEOMETRICCONSTRAINTSIDENTIFICATIONOFCONSTRAINTSINMODELSOFMECHANICALASSEMBLIESISNECESSARYFORSIMULATIONSOFTHEEFFECTSOFDIMENSIONANDTOLERANCECHANGESTHISPAPERPRESENTSTECHNIQUESFORTHEAUTOMATICIDENTIFICATIONOFSUCHCONSTRAINTSINCOMPUTERMODELSOFTHREEDIMENSIONALASSEMBLIESWITHNONORTHOGONALCONTACTSBETWEENCOMPONENTSURFACESANDKINEMATICJOINTSTHEAPPROACHRELIESONAGRAPHBASEDREPRESENTATIONOFTHEASSEMBLYSEARCHALGORITHMSFORIDENTIFYINGASSEMBLYCONSTRAINTSINTHISGRAPHAREPRESENTED?1998ELSEVIERSCIENCELTDALLRIGHTSRESERVEDKEYWORDSASSEMBLYMODELING,MECHANICALASSEMBLIES,CONSTRAINTSINTRODUCTIONTHEIDENTIFICATIONANDSOLUTIONOFCONSTRAINTRELATIONSHIPSBETWEENCOMPONENTDIMENSIONSINMECHANICALASSEMBLIESISASIGNIFICANTPROBLEMINCOMPUTERAIDEDDESIGNPHYSICALCONTACTSBETWEENCOMPONENTSCREATECONSTRAINTSONTHERELATIVEPOSITIONOFTHECOMPONENTS,THEIRNOMINALDIMENSIONS,ANDTHETOLERANCESONTHOSEDIMENSIONSTHENUMBEROFSUCHCONSTRAINTSCANBELARGEEVENFORRELATIVELYSIMPLEASSEMBLIESSPECIALIZEDTECHNIQUESARENEEDEDTOACCOUNTFORSUCHGEOMETRICCONSTRAINTSTHECONSTRAINTSIDENTIFIEDCANBEUSEDTO1PROVIDEDIMENSIONSENSITIVITYFEEDBACKTOTHEDESIGNER2IDENTIFYRELATIONSHIPSTHATTHEDESIGNERMAYNOTRECOGNIZEORFULLYCOMPREHEND3ALLOWFORREDUCTIONOFDESIGNTIMEBYMAKINGITEASIERTOMODIFYTHEEXISTINGDESIGNAND4BEUSEFULFORBOTHTOPDOWNANDBOTTOMUPDESIGNMODESTOPDOWNASSEMBLYDESIGNSYSTEMSCREATEFUNCTIONALCONSTRAINTSWHICHTHEGEOMETRYMUSTSATISFYBOTTOMUPASSEMBLYDESIGNUSESTHEGEOMETRYTODEFINECONSTRAINTSONTHEDESIGN’SBEHAVIORBOTHAPPROACHESRESULTINCONSTRAINTSTHATARISEFROMCONTACTSANDCONNECTIONSBETWEENTHECOMPONENTSOFTHEASSEMBLY,ANDINEITHERAPPROACHITISNECESSARYTOMAINTAINTHECONSISTENCYOFTHEASSEMBLYTHROUGHOUTDESIGNCHANGESCOMPUTERSUPPORTFORINTEGRATEDDESIGNOFPARTSANDASSEMBLIESREQUIRESCONSTRAINTIDENTIFICATIONANDMANAGEMENTTHEREARETWOCATEGORIESOFGEOMETRICASSEMBLYRELATIONSHIPSMATINGCONDITIONSANDKINEMATICJOINTSINGENERALTERMS,AMATINGCONDITIONISAGEOMETRICRELATIONSHIPBETWEENTWOORMORECOMPONENTSTHATHASSIGNIFICANCEINTHEDESIGNORFABRICATIONPROCESSESMATINGCONDITIONSINCLUDERELATIONSHIPSWHICHINVOLVECONTACTBETWEENPARTS,ASWELLASRELATIONSHIPSINWHICHTWOPARTSDONOTHAVECONTACT,SUCHASCLEARANCECONDITIONSTHEDISTINCTIONBETWEENMATINGCONDITIONSANDKINEMATICJOINTSISTHATTHEGEOMETRICRELATIONSHIPOFAMATINGCONDITIONISSTATICAMATINGCONDITIONDEFINESARELATIONSHIPBETWEENCOMPONENTSTHATMAYNOTHOLDIFCHANGESOCCURINTHEDIMENSIONSOFTHECOMPONENTSINCONTRAST,AKINEMATICJOINTISAGEOMETRICRELATIONSHIPBETWEENTWOCOMPONENTSTHATALLOWSRELATIVEMOTIONANDHOLDSDESPITECHANGESINTHECOMPONENT’SDIMENSIONSTHEKINEMATICJOINTISAFUNCTIONALSPECIFICATION,BUTTHEMATINGCONDITIONISNOTEXAMPLESOFKINEMATICJOINTSARETHEREVOLUTEJOINTANDTHEPRISMATICJOINTAMATINGSURFACEISASURFACEONACOMPONENTTHATISINVOLVEDINAMATINGCONDITIONTHEMATINGSURFACESOFACOMPONENT,OROFAGROUPOFCOMPONENTSCONNECTEDBYKINEMATICJOINTS,CANRESTRICTTHERANGEOFVALUESOFTHECOMPONENTS’DIMENSIONSANDTHEDEGREESOFFREEDOMDOFOFTHEKINEMATICJOINTSASETOFCOMPONENTSTHATARERELATEDBYASETOFKINEMATICJOINTSISAKINEMATICGROUP,ANDASINGLECOMPONENTORAKINEMATICGROUPISACONSTRAINEDGROUPTHETERMCONSTRAINEDGROUPINFERSTHEREAREGEOMETRICRELATIONSHIPSBETWEENTHEMATINGSURFACESTHAT1ARENOTIDENTIFIABLEASASSEMBLYCONSTRAINTS,AND2AREUSEFULINIDENTIFYINGANDFORMULATINGASSEMBLYCONSTRAINTSFORASINGLECOMPONENT,THESEGEOMETRICRELATIONSHIPSARETHECOMPONENT’SDIMENSIONSANDTOLERANCESFORAKINEMATICGROUP,THEYALSOINCLUDETHEDOFOFTHEKINEMATICJOINTSANASSEMBLYCONSTRAINTISCREATEDBYMATINGCONDITIONSCOMPUTERAIDEDDESIGN,VOL30,NO9,PP715–726,1998?1998ELSEVIERSCIENCELTDALLRIGHTSRESERVEDPRINTEDINGREATBRITAIN00104485/98/1900000PIIS0010448598000268715TOWHOMCORRESPONDENCESHOULDBEADDRESSEDTEL7654945720FAX7654940811EMAILDAVEECNPURDUEEDU?STRUCTURALDYNAMICRESEARCHCORPORATION,MILFORD,OH451502789,USA?SCHOOLOFMECHANICALENGINEERING,PURDUEUNIVERSITY,WESTLAFAYETTE,IN479061288,USAPAPERRECEIVED5APRIL1996REVISED10MARCH1998ACCEPTED20MARCH1998MATINGSURFACESOFCOMPONENTACANCOMBINETORESTRICTITSMOTIONINTHETWODIMENSIONALSUBSPACEOFE3DEFINEDBYTHEPLANEOFTHEPAPERNOPAIROFMATINGSURFACESCANRESTRICTTHETRANSLATIONOFTHECOMPONENTINTHISSUBSPACESUCHASETOFMATINGSURFACESISDEFINEDASAPHYSICALLYCONSTRAININGFACESETPCFSTHEPCFSISSIGNIFICANTBECAUSEWITHOUTSUCHASETOFMATINGSURFACES,THECOMPONENTDIMENSIONSINTHECONSTRAINEDGROUPCOULDBECHANGEDWITHOUTREGARDTOTHEOTHERCOMPONENTSINTHEASSEMBLYTHECONSTRAINEDGROUPCOORDINATESYSTEMCOULDTRANSLATEDUETOCHANGESINTHEDIMENSIONSOFANYOFTHECOMPONENTSINTHEASSEMBLYTHENEXTSECTIONDESCRIBESAGENERALMATHEMATICALDEFINITIONOFPCFSTHATWILLENABLETHEIDENTIFICATIONOFGENERALTRANSLATIONALCONSTRAINTSONCOMPONENTSCHARACTERISTICVECTORSPACEEACHMATINGSURFACEHASACHARACTERISTICVECTORCVSPACEOFDIRECTIONSFORWHICHITPREVENTSTRANSLATIONOFTHEASSOCIATEDCONSTRAINEDGROUPFIGURE4DEMONSTRATESTHECVSPACESFORSEVERALTYPESOFMATINGSURFACESINFIGURE4A,THEPLANARSURFACEMATINGAGAINSTANOTHERSURFACEHASACVSPACEDEFINEDBYA?QANDA?0INWHICH?QHASTHESAMEDIRECTIONASTHESURFACENORMALTHECYLINDEROFFIGURE4BINACYLINDRICALFITSCONDITIONHASACVSPACEALSODEFINEDBYA?Q,A?0,WHERE?QISDERIVEDFROM?Q?C?QXTD?QY2C2TD2?13?QXAND?QYAREORTHONORMALVECTORSPERPENDICULARTOTHEAXISOFTHECYLINDERTHECVSPACEFORASPHERICALMATINGSURFACEISDEFINEDSIMILARLYINFIGURE4C?QX,?QYAND?QZORTHOGONALVECTORSANDTHECVSPACEISTHENGIVENBYA?Q,A?0,WHERE?Q?C?QXTD?QYTE?QZ4C2TD2TE2?15THECVSPACEFORAFREEFORMPARAMETRICSURFACEISDEFINEDBYTHESURFACENORMALANDTHESURFACEPARAMETERSTHECVSPACEISTHENDEFINEDBYA?Q,A?0,WHERE?Q??QS,TT??SR???TRK??SR???TRK6INWHICHR?RS,TISTHEPOSITIONOFAPOINTONTHESURFACEFORTHEGIVENPARAMETERVALUESALGEBRAICIDENTIFICATIONOFCOMPONENTSPATIALCONSTRAINTSTHESECVSPACESCANNOWBEUSEDTOIDENTIFYTHEPCFSFORCONSTRAINEDGROUPSTWOMATINGSURFACESAREREQUIREDTORESTRICTTHECONSTRAINEDGROUP’STRANSLATIONINAONEDIMENSIONALSUBSPACEOFE3THATIS,THECHARACTERISTICVECTORSPACESFORTHEPAIROFMATINGSURFACESMUSTBOTHSPAN,ANDBERESTRICTEDTO,AONEDIMENSIONALSUBSPACETHISREQUIREMENTCANBEEXPRESSEDASA1?Q1TA2?Q2?07A1?08A2?09IFEQNS7–9HAVEASOLUTIONFORTHEVARIABLESA1ANDA2THENMATINGSURFACES1AND2FORMAONEDIMENSIONALCONSTRAINTTHEIMPORTANTPHYSICALASPECTOFTHERELATIONSHIPINEQN7ISTHAT,IFTHERELATIONSHIPDOESNOTHOLD,THEMATINGSURFACESARENOTMUTUALLYCONSTRAININGANYOFTHECVSPACESCANBEINSERTEDINTOTHERELATIONSHIPOFEQN7PROVIDEDTHEAUXILIARYCONSTRAINTSOFEQN3OREQN5AREINCLUDEDWHENAPPROPRIATEINTHECASEOFAONEDIMENSIONALCONSTRAINT,EQNS7–9HAVEASOLUTIONONLYIF?Q1AND?Q2CANBEMADETOBEEQUALANDOPPOSITETHEGENERALIZATIONOFEQNS7–9TOTWODIMENSIONALCONSTRAINTSCANBEDERIVEDBYEXAMININGFIGURE3FORTHETRANSLATIONOFTHECONSTRAINEDGROUPOFPARTATOBECOMPLETELYRESTRAINED,ITMUSTNOTBEPOSSIBLETOIMPOSEADISPLACEMENTOFAI?0ALONGANYOFTHECHARACTERISTICVECTORS?QIWITHOUTTHISRESULTINGINADISPLACEMENTOFAJ?0ALONGSOMEOTHERCHARACTERISTICVECTOR?QJTHISREQUIREMENTCANBESUMMARIZEDBYTHEFOLLOWINGRELATIONSHIPSA1?Q1TA2?Q2TA3?Q3?010AJ?011A2?012A3?013AGAIN,ANYOFTHECHARACTERISTICVECTORSPACESCANBESUBSTITUTEDINTOEQN10WITHTHESTIPULATIONTHATALLAUXILIARYRELATIONS,SUCHASTHOSEOFEQN3ANDEQN5,AREINCLUDEDTHERELATIONSHIPSBETWEENTHECHARACTERISTICVECTORSCANGEOMETRICCONSTRAINTSINMECHANICALASSEMBLIESSHMULLINSANDDCANDERSON717FIGURE3TWODIMENSIONALPHYSICALLYCONSTRAININGFACESETPCFSFIGURE4CHARACTERISTICVECTORSPACEFORSEVERALSURFACETYPES

簡介：EFFICIENTPLANNINGOFSUBSTATIONAUTOMATIONSYSTEMCABLESTHANIKESAVANSIVANTHIANDJANPOLANDABBSWITZERLANDLTD,CORPORATERESEARCH,SEGELHOFSTRASSE1K,5405,BADEND¨ATTWIL,AARGAU,SWITZERLANDABSTRACTTHEMANUALSELECTIONANDASSIGNMENTOFAPPROPRIATECABLESTOTHEINTERCONNECTIONSBETWEENTHEDEVICESOFASUBSTATIONAUTOMATIONSYSTEMISAMAJORCOSTFACTORINSUBSTATIONAUTOMATIONSYSTEMDESIGNTHISPAPERDISCUSSESABOUTTHEMODELINGOFTHESUBSTATIONAUTOMATIONSYSTEMCABLEPLANNINGASANINTEGERLINEAROPTIMIZATIONPROBLEMTOGENERATEANEFFICIENTCABLEPLANFORSUBSTATIONAUTOMATIONSYSTEMS1INTRODUCTIONCABLINGBETWEENDIFFERENTDEVICESOFASUBSTATIONAUTOMATIONSYSTEMSASISAMAJORCOSTFACTORINTHESASDESIGNPROCESSUSUALLYCOMPUTERAIDEDDESIGNSOFTWAREISUSEDTOCREATETHEDESIGNTEMPLATESOFSASDEVICESANDTHEIRINTERCONNECTIONSTHEDESIGNTEMPLATESARETHENINSTANTIATEDINASASPROJECTANDTHECABLESAREMANUALLYASSIGNEDTOTHECONNECTIONSTHESELECTIONANDASSIGNMENTOFCABLESTOCONNECTIONSMUSTFOLLOWCERTAINENGINEERINGRULESTHISENGINEERINGPROCESSISUSUALLYTIMECONSUMINGANDCANCAUSEENGINEERINGERRORS,THEREBYINCREASINGTHEENGINEERINGCOSTAPPARENTLY,THESASCABLEPLANNINGISRELATEDTOTHEWELLKNOWNBINPACKINGPROBLEMTHESASCABLEPLANNINGCANBEFORMULATEDASANINTEGERLINEAROPTIMIZATIONPROBLEMWITHTHECABLEENGINEERINGRULESEXPRESSEDASASETOFLINEARCONSTRAINTSANDACOSTOBJECTIVEFORMINIMIZINGTHETOTALCABLECOSTTHISPAPERDESCRIBESTHEFORMULATIONOFSASCABLEPLANNINGPROBLEMASANINTEGERLINEAROPTIMIZATIONPROBLEMANDPRESENTSTHERESULTSFORSOMEREPRESENTATIVETESTCASESTOTHEBESTOFTHEAUTHORS’KNOWLEDGETHEWORKISTHEFIRSTOFTHEKINDTOSTUDYSASCABLEPLANNINGTHEPAPERISORGANIZEDASFOLLOWSSECTION2PRESENTSANOVERVIEWOFTHESASCABLEPLANNINGPROCESSSECTION3EXPRESSESTHESASCABLEPLANNINGPROBLEMASANINTEGERLINEAROPTIMIZATIONPROBLEMTHERESULTSOBTAINEDBYSOLVINGTHEOPTIMIZATIONPROBLEMUSINGSOMESOLVERSISPRESENTEDINSECTION4SECTION5DRAWSSOMECONCLUSIONSOFTHISWORK2SASCABLEPLANNINGTHESASCABLEPLANNINGBEGINSAFTERTHESYSTEMDESIGNPHASEOFASASPROJECTTHESASCABLEPLANNINGISATPRESENTDONEMANUALLYBYCOMPUTERAIDEDDESIGNTACHTERBERGANDJCBECKEDSCPAIOR2011,LNCS6697,PP210–214,2011CSPRINGERVERLAGBERLINHEIDELBERG2011EFFICIENTPLANNINGOFSUBSTATIONAUTOMATIONSYSTEMCABLES211TOTALNUMBEROFCONNECTIONS,ANDK{1,2,3,,M}212TSIVANTHIANDJPOLANDREPRESENTTHESETOFALLCABLETYPES,WHEREMISTHETOTALNUMBEROFCABLETYPESINASUBPROBLEMINACABLEINSTANCE,THERECANBEONEORMORECONNECTIONSANDWEREFERTOTHECONNECTIONWITHLOWESTINDEXAMONGALLCONNECTIONSINTHECABLEINSTANCEASTHELEADERANDTHEOTHERCONNECTIONSASTHEFOLLOWERSTHISIMPLIESTHATALLCONNECTIONSEXCEPTTHEFIRSTCONNECTIONINCCANEITHERBEALEADERORFOLLOWERMOREOVER,BASEDONTHESIGNALRULESASETOFCONNECTIONPAIRSXCANBEDERIVEDWHEREEACHI,?I∈XREPRESENTSTHECONNECTIONSIAND?ITHATMUSTNOTBEASSIGNEDTOTHESAMECABLELETˉCBETHESETOFCONNECTIONPAIRSI,?IWHEREI,?I∈C,I?I,I,?I/∈XWEINTRODUCETHEFOLLOWINGBINARYVARIABLEXI,?I,WHEREI,?I∈ˉC,WHICHWHENTRUEIMPLIESTHATCONNECTIONIISAFOLLOWEROFALEADER?I1CIIWHEREORXII??,,10,SIMILARLY,BASEDONTHECABLERULESASETOFCONNECTIONCABLEPAIRSYCANBEDERIVEDWHEREEACHI,J∈YIMPLIESTHATCABLETYPEJISNOTALLOWEDFORCONNECTIONILETˉKBETHESETOFCONNECTIONCABLEPAIRSI,J,WHEREI∈C,J∈K,I,J/∈YWEINTRODUCETHEFOLLOWINGBINARYVARIABLEYI,J,WHEREI,J∈ˉK,WHICHWHENTRUEIMPLIESTHATTHELEADERIISASSIGNEDTOANINSTANCEOFCABLETYPEJ2_,,,10KJIWHEREORXJI??TABLE1ILLUSTRATESALLBINARYVARIABLESCORRESPONDINGTOTHEEXAMPLESHOWNINFIGURE1FORTHECASEWITHTWOCABLETYPESK1ANDK2ITISASSUMEDTHATCONNECTIONSC1ANDC3CANNOTBEASSIGNEDTOTHESAMECABLEANDK1ISNOTANALLOWEDCABLETYPEFORCONNECTIONC3ASMENTIONEDBEFOREALLCONNECTIONSEXCEPTTHEFIRSTCONNECTION,WHICHMUSTBEALEADER,CANEITHERBEALEADERORFOLLOWERTHISISENSUREDBYTHEFOLLOWINGCONSTRAINT（3）CIYXKJIKJJICIIII?????????,1,,,,,ACONNECTIONWHICHISALEADERINACABLECANNOTBEAFOLLOWEROFALEADERINANOTHERCABLETHISISEXPRESSEDBYTHEFOLLOWINGCONSTRAINT（4）_,,,,,1_CIIXXCIIIIII??????ANIMPLICITCONSTRAINTOFTHECABLEPLANNINGPROBLEMISTHECAPACITYCONSTRAINTWHICHIMPLIESTHATTHENUMBEROFCONNECTIONSASSIGNEDTOACABLEMUSTBELESSTABLE1BINARYVARIABLESCORRESPONDINGTOFIGURE1EXAMPLE

簡介：附錄附錄B英文原文及翻譯英文原文及翻譯

簡介：ARXIV10035062V1PHYSICSGENPH26MAR2010AUTOMATICPOLISHINGPROCESSOFPLASTICINJECTIONMOLDSONA5AXISMILLINGCENTERJOURNALOFMATERIALSPROCESSINGTECHNOLOGYXAVIERPESSOLES,CHRISTOPHETOURNIERLURPA,ENSCACHAN,61AVDUPDTWILSON,94230CACHAN,FRANCECHRISTOPHETOURNIERLURPAENSCACHANFR,TEL33147402996,FAX33147402211ABSTRACTTHEPLASTICINJECTIONMOLDMANUFACTURINGPROCESSINCLUDESPOLISHINGOPERATIONSWHENSURFACEROUGHNESSISCRITICALORMIRROREFFECTISREQUIREDTOPRODUCETRANSPARENTPARTSTHISPOLISHINGOPERATIONISMAINLYCARRIEDOUTMANUALLYBYSKILLEDWORKERSOFSUBCONTRACTORCOMPANIESINTHISPAPER,WEPROPOSEANAUTOMATICPOLISHINGTECHNIQUEONA5AXISMILLINGCENTERINORDERTOUSETHESAMEMEANSOFPRODUCTIONFROMMACHININGTOPOLISHINGANDREDUCETHECOSTSWEDEVELOPSPECIALALGORITHMSTOCOMPUTE5AXISCUTTERLOCATIONSONFREEFORMCAVITIESINORDERTOIMITATETHESKILLSOFTHEWORKERSTHESEAREBASEDONBOTHFILLINGCURVESANDTROCHOIDALCURVESTHEPOLISHINGFORCEISENSUREDBYTHECOMPLIANCEOFTHEPASSIVETOOLITSELFANDSETUPBYCALIBRATIONBETWEENDISPLACEMENTANDFORCEBASEDONAFORCESENSORTHECOMPLIANCEOFTHETOOLHELPSTOAVOIDKINEMATICALERROREFFECTSONTHEPARTDURING5AXISTOOLMOVEMENTSTHEEFFECTIVENESSOFTHEMETHODINTERMSOFTHESURFACEROUGHNESSQUALITYANDTHESIMPLICITYOFIMPLEMENTATIONISSHOWNTHROUGHEXPERIMENTSONA5AXISMACHININGCENTERWITHAROTARYANDTILTTABLEKEYWORDSAUTOMATICPOLISHING,5AXISMILLINGCENTER,MIRROREFFECT,SURFACEROUGHNESS,HILBERT’SCURVES,TROCHOIDALCURVES1SURFACEROUGHNESSPARAMETERSRAARITHMETICAVERAGEDEVIATIONOFTHESURFACE2DSAARITHMETICALMEANHEIGHTOFTHESURFACE3DSQROOTMEANSQUAREDEVIATIONOFTHESURFACESSKSKEWNESSOFTOPOGRAPHYHEIGHTDISTRIBUTIONSKUKURTOSISOFTOPOGRAPHYHEIGHTDISTRIBUTION3