Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/98234
標題: 一種利用Q-learning調整PID控制器的方法
A Method of Adjusting PID Controller Parameter Based on Q-learning
作者: 蕭淳方
Chun-Fang Hsiao
關鍵字: PID控制器
PID設定
Q-Learning
禁忌演算法
PID controller
PID setting
Q-Learning
TABU Search
引用: [1]MBA智库百科,'試誤法'http://wiki.mbalib.com/zh-tw/%E8%AF%95%E8%AF%AF%E6%B3%95 [2]KarlJ.Astrom,ToreHagglund,AdvancedPIDControl,ISA-Instrumentation,Systems,andAutomationSociety,pp.1-229,2016 [3]Wikipedia,'強健性'https://zh.wikipedia.org/wiki/%E9%B2%81%E6%A3%92%E6%80%A7 [4]J.G.ZieglerandN.B.Nichols,'Optimumsettingsforautomaticcontrollers,'Tran.ASME,Vol.64,pp.759-768,1942. [5]Chang-ChienHangandLisengCao,'Improvementoftransientresponsebymeansofvariablesetpointweighting,'IEEETransactionsonIndustrialElectronics,Vol.43,pp.477-484,1996. [6]IKaya,'Controllerdesignforintegratingprocessesusinguser-specifiedgainandphasemarginspecificationsandtwodegree-of-freedomIMCstructure,'IEEEConferenceonControlApplications,Vol.2,2003,pp.898-902. [7]M.A.DahlehandI.J.Diaz-Bobillo,ControlofUncertainSystems:ALinearProgrammingApproach,PrenticeHall,1995. [8]Hyun-JoonCho,Kwang-BoChoandBo-HyeunWang,'Fuzzy-PIDhybridcontrol:Automaticrulegenerationusinggeneticalgorithms,'FuzzySetsandSystems,Vol.92,pp.305-316,1997. [9]Ming-TzuHo,A.Datta,andS.P.Bhattacharyya,'AlinearprogrammingcharacterizationofallstabilizingPIDcontrollers,'ProceedingsofAmericanControlConference,Vol.6,pp.3922-3928,1997. [10]N.Tan,I.KayaandD.P.Atherton,'ComputationofstabilizingPIandPIDcontrollers,'IEEEConferenceonControlApplications,Vol.2,pp.876-881,2003. [11]K.J.Åström,andT.Hägglund,'ThefutureofPIDcontrol.'Controlengineeringpractice9.11:1163-1175,2001. [12]Wikipedia,'PID控制器'https://zh.wikipedia.org/wiki/PID%E6%8E%A7%E5%88%B6%E5%99%A8 [13]TerryBartelt.《工業控制電子學》(第2版).清華大學出版社(原書由Delmar出版).2005年.ISBN7-302-09149-8 [14]NI國家儀器,'Chien-Hrones-ReswickAutotuningMethod'http://zone.ni.com/reference/en-XX/help/370401J-01/lvpidmain/chienforms/ [15]ControlNotes,'Cohen-Coontuningrules'http://blog.opticontrols.com/archives/383 [16]徐韶鴻,'應用改良型粒子群演算法於分數階系統之分數階PID控制器設計',台灣,2015. [17]F.Glover,M.Laguna,andR.Marti,Principlesoftabusearch,HandbookofApproximationAlgorithmsandMetaheuristics.Taylor&FrancisGroup,LLC,2007.
摘要: 隨著科技發展,許多工作已遭機械取代,在這種情況下而如何準確進行控制機械顯得特別重要,於是PID控制器便應運而生了。雖然PID控制器為廣泛的應用在現今控制系統中,但傳統PID控制器的參數調整,常需要工程師憑經驗調適,不僅耗時且調整門檻高,若遇到非線性的系統,更是難以達到最佳控制需求。因此,希望有一種自動調整PID控制器參數的技術。 於是本文研究的方向將針對利用Q-Learning演算法搭配禁忌搜尋法來學習PID控制器參數調整,讓沒有調整過PID控制器的工程師們,也可以透過本方法來找出最佳控制需求的PID控制器參數。透過訓練出之數據後,投入實體機具,來達到控制的效果。本論文中的實驗結果發現,照此方法投入溫控環境是有達到控制溫度的效果。
With advance in technology , most manual work has been replaced by mechanically ,thus accurate controlling machine is significant in this situation ,in this way, PID controller has been created . Although the PID controller widely used in control system nowadays, but traditional PID control system parameter adjustment need experienced engineer ,in this way, adjustment parameter not only time-consuming but also difficult, especially in nonlinear system .Therefore, hope to propose automatic technology adjust PID control system parameter. Research in this paper mean to using Q-Learning with TABU search algorithm for learning PID control system adjustment. For inexperienced engineer also can adjust PID control system parameter .Using this research to training parameter and put it in actual environment .In this search, we prove the method which can be used in temperature control system by experimental result.
URI: http://hdl.handle.net/11455/98234
文章公開時間: 10000-01-01
Appears in Collections:資訊科學與工程學系所

文件中的檔案:

取得全文請前往華藝線上圖書館

Show full item record
 
TAIR Related Article
 
Citations:


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.