本文選題：劣化系統(tǒng) + 維修策略��； 參考：《合肥工業(yè)大學(xué)》2011年碩士論文

【摘要】：工業(yè)生產(chǎn)中,受到運(yùn)行時(shí)間和所處環(huán)境的影響,生產(chǎn)系統(tǒng)的狀態(tài)不斷劣化,工作效率和性能都逐漸下降。當(dāng)下降到無(wú)法滿足工作要求時(shí),即使系統(tǒng)還能工作,仍將其視為失效,系統(tǒng)失效會(huì)造成經(jīng)濟(jì)上的巨大損失。事前維修是指利用一種或一系列的維修作業(yè),發(fā)現(xiàn)或排除某一隱蔽或潛在故障,使系統(tǒng)保持在良好的工作狀態(tài),避免系統(tǒng)失效,這對(duì)于減少生產(chǎn)成本以及工業(yè)生產(chǎn)有著重要的影響。因此,如何對(duì)生產(chǎn)系統(tǒng)的維修進(jìn)行調(diào)配,避免系統(tǒng)在一個(gè)生產(chǎn)成本較高的狀態(tài)下運(yùn)行,以及對(duì)提高系統(tǒng)的可靠性和安全性是一個(gè)重要的研究課題。 論文以強(qiáng)化學(xué)習(xí)為基礎(chǔ),首先針對(duì)離散狀態(tài)下的劣化系統(tǒng)維修問(wèn)題,建立了連續(xù)時(shí)間的半馬爾可夫決策過(guò)程(Semi-Markov Decision Process, SMDP)模型。為了避免結(jié)果陷入局部最優(yōu)值,使用了Q學(xué)習(xí)與模擬退火(Simulated Annealing, SA)相結(jié)合的算法對(duì)該問(wèn)題進(jìn)行求解,得到系統(tǒng)較優(yōu)的維修策略。通過(guò)仿真得出平均和折扣性能準(zhǔn)則下的優(yōu)化結(jié)果,并討論了檢測(cè)間隔時(shí)間對(duì)結(jié)果的影響。 同時(shí),論文還考慮了部分可觀的劣化系統(tǒng),也即檢測(cè)存在誤差觀測(cè)者不能完全確定系統(tǒng)的狀態(tài),而只能通過(guò)不完整的信息來(lái)對(duì)系統(tǒng)進(jìn)行決策的情況,針對(duì)離散狀態(tài)連續(xù)時(shí)間下的問(wèn)題建立了部分可觀半馬爾可夫決策過(guò)程(Partially Observed Semi-Markov Decision Process,POSMDP)模型,利用了強(qiáng)化學(xué)習(xí)中的Sara (λ)學(xué)習(xí)算法以及NSM算法,分別從無(wú)記憶和基于記憶的角度來(lái)對(duì)問(wèn)題進(jìn)行求解,得到了在平均性能準(zhǔn)則下的優(yōu)化結(jié)果。同時(shí)對(duì)檢測(cè)間隔與平均代價(jià)之間的影響進(jìn)行了討論,與完全可觀下的結(jié)果一致。最后,論文還對(duì)NSM算法中參數(shù)k的取值進(jìn)行了討論,與實(shí)際情況相符合。
[Abstract]:Due to the influence of running time and environment, the state of the production system is deteriorating, and the efficiency and performance of the production system decrease gradually. Even if the system can still work, it can still be regarded as invalid, and the failure of the system will cause huge economic losses. Prior maintenance refers to the use of one or a series of maintenance operations to detect or eliminate a hidden or potential fault, so as to keep the system in good working condition and to avoid system failure. This has an important impact on reducing production costs and industrial production. Therefore, how to adjust the maintenance of the production system, avoid the system running under a high production cost, and improve the reliability and safety of the system is an important research topic. Based on reinforcement learning, a semi-Markov Decision Process, SMDP) model of continuous time semi-Markov decision process is established for the maintenance of degraded systems in discrete state. In order to avoid the result falling into the local optimal value, the problem is solved by the combination of Q learning and simulated annealing algorithm (SA), and the optimal maintenance strategy of the system is obtained. The optimization results under average and discounted performance criteria are obtained by simulation, and the effect of detection interval on the results is discussed. At the same time, the paper also considers some considerable deterioration systems, that is, the detection of error observers can not fully determine the state of the system, but only through incomplete information to make decisions on the system. A partial observable Observed Semi-Markov Decision process POSMDP model is established for discrete state continuous time problem. Sara (位) learning algorithm and NSM algorithm are used in reinforcement learning. The problem is solved from the point of view of memoryless and memoryless, and the optimization results under the average performance criterion are obtained. At the same time, the effect of detection interval and average cost is discussed, which is in agreement with the results obtained under completely observable conditions. Finally, the parameter k in NSM algorithm is discussed, which is in accordance with the actual situation.
【學(xué)位授予單位】：合肥工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2011
【分類號(hào)】：TH17

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 程志君;郭波;;多部件系統(tǒng)機(jī)會(huì)維修優(yōu)化模型[J];工業(yè)工程;2007年05期

2 修國(guó)明;張積濱;潘啟樹;;基于實(shí)例的POMDP問(wèn)題的近似求解[J];計(jì)算機(jī)工程與應(yīng)用;2008年29期

3 王學(xué)寧,賀漢根,徐昕;求解部分可觀測(cè)馬氏決策過(guò)程的強(qiáng)化學(xué)習(xí)算法[J];控制與決策;2004年11期

4 曹希仁;;離散事件動(dòng)態(tài)系統(tǒng)[J];自動(dòng)化學(xué)報(bào);1985年04期

5 程志君;郭波;;基于半Markov決策過(guò)程的劣化系統(tǒng)檢測(cè)與維修優(yōu)化模型[J];自動(dòng)化學(xué)報(bào);2007年10期

6 仵博;吳敏;;部分可觀察馬爾可夫決策過(guò)程研究進(jìn)展[J];計(jì)算機(jī)工程與設(shè)計(jì);2007年09期

7 趙喜林;許興華;;現(xiàn)代制造系統(tǒng)維修策略研究[J];新技術(shù)新工藝;2006年01期

8 程志君;郭波;;連續(xù)劣化系統(tǒng)的最佳檢測(cè)與維修策略分析[J];系統(tǒng)工程與電子技術(shù);2008年01期

9 王文義,張洪芬;維修策略的概念、方法和模型(Ⅰ)[J];運(yùn)籌與管理;1997年02期

，

本文編號(hào)：1921910