本文選題：可靠性 + 可靠度��； 參考：《西南大學(xué)》2014年碩士論文

【摘要】：現(xiàn)代社會工業(yè)生產(chǎn)的大規(guī)模發(fā)展,為人類提供了豐富的物質(zhì)生活產(chǎn)品,但同時也給人類帶來了風(fēng)險。每年都有數(shù)十萬人死于工業(yè)生產(chǎn)中的安全事故。工業(yè)安全事故頻發(fā)的主要原因是工業(yè)上的系統(tǒng)設(shè)備的可靠性達(dá)不到要求造成的。為了減少安全事故的發(fā)生就必須工業(yè)系統(tǒng)的可靠性,而要提高系統(tǒng)的可靠性首先就要對系統(tǒng)進(jìn)行準(zhǔn)確地評估,然后才能對現(xiàn)有系統(tǒng)采取相應(yīng)的措施來提高系統(tǒng)的可靠性。因此對現(xiàn)有系統(tǒng)進(jìn)行可靠性評估是非常重要,對可靠性評估方法進(jìn)行研究就非常有意義�，F(xiàn)有文獻(xiàn)中有很多針對可靠性評估方法進(jìn)行了研究,可靠性評估的方法主要由故障樹分析法、可靠性框圖法以及模型方法。故障樹分析法和可靠性框圖法能夠?qū)φ麄�系統(tǒng)進(jìn)行可靠性分析,但是必須是針對一個具體的系統(tǒng)。而工業(yè)系統(tǒng)中提高系統(tǒng)可靠性的方法就是給系統(tǒng)做冗余,模型分析法就是用來分析這樣的系統(tǒng)的可靠性,它針對的是一類系統(tǒng)。因此基本上的研究都是基于模型分析法。本文的研究基于模型分析法展開的。模型分析的方法中的模型主要由串聯(lián)系統(tǒng)模型、并聯(lián)系統(tǒng)模型、k/n表決系統(tǒng)模型以及Markov模型�；旧系娜哂嘞到y(tǒng)都可以通過這些模型來對其進(jìn)行可靠性評估。但是有一類特殊的冗余系統(tǒng)——系統(tǒng)中的存在中心單元,另外的冗余單元只是輔助這個中心單元(就是單個的冗余單元沒有對中心單元的全部功能做冗余,只是冗余了其中的部分功能,即整個冗余系統(tǒng)中單元的重要性不一樣)——無法用現(xiàn)有的模型對其進(jìn)行可靠性評估。本文針對模型分析法中存在的缺陷,在k/n表決系統(tǒng)模型的基礎(chǔ)上引入權(quán)重,提出了基于權(quán)重的表決系統(tǒng)模型。該模型能夠準(zhǔn)確評估上述的特殊系統(tǒng)的可靠性,彌補(bǔ)了現(xiàn)有模型分析法中的不足,完善了模型分析法。 本文首先介紹了可靠性的基本概念和可靠性模型分析法中的幾個模型,具體介紹了k/n表決系統(tǒng)模型；其次基于模型分析法中現(xiàn)有模型的不足,在k/n表決系統(tǒng)模型的基礎(chǔ)上進(jìn)行改進(jìn),提出了權(quán)重表決系統(tǒng)模型；然后舉例分析權(quán)重表決系統(tǒng)模型在特殊系統(tǒng)中的應(yīng)用；接著通過實(shí)驗(yàn)實(shí)現(xiàn)了權(quán)重表決系統(tǒng)模型,并開發(fā)了可以用來評估系統(tǒng)可靠性的軟件；最后對現(xiàn)有的工作進(jìn)行了總結(jié),并對以后的工作進(jìn)行了展望。
[Abstract]:The large-scale development of industrial production in modern society provides human beings with abundant material life products, but also brings risks to human beings. Hundreds of thousands of people die every year from safety accidents in industrial production. The main reason for frequent industrial safety accidents is that the reliability of system equipment in industry is not up to requirements. In order to reduce the occurrence of safety accidents, it is necessary to improve the reliability of the industrial system. To improve the reliability of the system, it is necessary to evaluate the system accurately, and then take corresponding measures to improve the reliability of the existing system. Therefore, it is very important to evaluate the reliability of existing systems, and it is very meaningful to study the reliability evaluation methods. There are many researches on reliability evaluation methods in the literature. The methods of reliability evaluation mainly include fault tree analysis method, reliability block diagram method and model method. The method of fault tree analysis and reliability block diagram can analyze the reliability of the whole system, but it must be aimed at a specific system. The method of improving the system reliability in industrial system is to make redundancy for the system. The model analysis method is used to analyze the reliability of such a system, which is aimed at a class of systems. Therefore, the basic research is based on model analysis. The research of this paper is based on the model analysis method. The models of model analysis are mainly composed of series system model, parallel system model, k / n voting system model and Markov model. These models can be used to evaluate the reliability of redundant systems. But there is a special kind of redundant system-the existence of central units in the system, and the other redundant units only assist this central unit (that is, a single redundant unit does not make redundancy for all the functions of the central unit. Only some of the functions are redundant, that is, the importance of the unit in the whole redundant system is different-it is impossible to evaluate the reliability of the redundant system by using the existing model. Aiming at the defects of the model analysis method, this paper introduces the weight on the basis of the KR / n voting system model, and puts forward the voting system model based on the weight. The model can accurately evaluate the reliability of the special system mentioned above, make up for the deficiency of the existing model analysis method, and perfect the model analysis method. This paper first introduces the basic concept of reliability and several models in reliability model analysis, and then introduces the KR / n voting system model, secondly, based on the shortcomings of existing models in model analysis, The weight voting system model is improved on the basis of the KR / n voting system model. Then the application of the weight voting system model in a special system is analyzed with an example. Then, the weight voting system model is realized through experiments. The software which can be used to evaluate the reliability of the system is developed. Finally, the existing work is summarized and the future work is prospected.
【學(xué)位授予單位】：西南大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TB114.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 馬平;呂鋒;杜海蓮;王瑞;牛成林;;多傳感器信息融合基本原理及應(yīng)用[J];控制工程;2006年01期

2 周芳;韓立巖;;多傳感器信息融合技術(shù)綜述[J];遙測遙控;2006年03期

3 何友,關(guān)欣,王國宏;多傳感器信息融合研究進(jìn)展與展望[J];宇航學(xué)報;2005年04期

，

本文編號：2046528