【正文】 tion of observers, etc. Since it requires accurately system model, this method is not economically feasible for the plicated devices in the practice. Pattern recognition conducts cluster description for a series of process or events. It is mainly divided into statistical method and language structure method. The fault diagnosis of equipments could be recognized as the pattern recognition process, that is to say, it recognizes the fault based on the extraction of fault characteristics. There are many mon recognition methods, including bayes category, distance function category, fuzzy diagnosis, fault tree analysis, grey theory diagnosis and so on. Recent years, some new technologies have been also applied in the field of the fault diagnosis of rotary machines, such as the bination of fuzzy set and neural work, the dynamic pattern recognition based on hidden markov model, etc. 外文翻譯原文及譯文 10 5. Research and Development of Fault Diagnosis Devices Fault diagnosis technology ultimately es down to the actual devices, and at present research and development of fault diagnosis devices is in the following two directions: (1) Portable vibration monitoring and diagnosis (including data collector system), and (2) Online condition monitoring and fault diagnosis system. Portable instrument is mainly adopted singlechip microputers to plete data acquisition, which has certain ability for signal analysis and fault diagnosis. Online monitoring and diagnosis system is usually equipped with sensors, data acquisition, alarm and interlock protection, condition monitoring subsystem, etc. And it is also fitted with rich signal analysis and diagnosis software. These software include America BENTLY Corporation 3300, 3500 and DM2020 systems, America Westinghouse Company PDS system, the 5911 system developed by ENTECK and IRD Company, Japan Mitsubishi MHM system, the Danish Bamp。 7: 157172. [6] DaneshiFar Z, Capolino GA, Henao H. Review of failures and condition monitoring in wind turbine generators. 19th International Conference on Electrical Machines. Rome, Italy。研究結(jié)果表明，基于智能信息融合的機(jī)械故障診斷專(zhuān)家系統(tǒng)與自我學(xué)習(xí)和自我更新能力，是機(jī)械設(shè)備狀態(tài)監(jiān)測(cè)和故障診斷未來(lái)研究的發(fā)展方向。在過(guò)去的幾十年里，已經(jīng)有許多研究者在此領(lǐng)域做了大量的工作 。 Xu 等人 [11]總結(jié)了旋轉(zhuǎn)機(jī)的常見(jiàn)故障。在某種意義上說(shuō)，傳統(tǒng)的信號(hào)分析，仍然是機(jī)械振動(dòng)信號(hào)分析和故障特征提取的主要方法。這項(xiàng)技術(shù)是指模型的建立，參數(shù)估計(jì)，狀態(tài)估計(jì)，應(yīng)用觀察員等。不管是傳統(tǒng)的還是先進(jìn)的故障診斷技術(shù)在各種應(yīng)用中都已經(jīng)取得了很大進(jìn)步，按照信息系統(tǒng)的觀點(diǎn)，每項(xiàng)技術(shù)都是故障診斷的組成部分，所有部分的有效的融合是最好實(shí)現(xiàn)條件監(jiān)控和故障診斷的保障。K 公司開(kāi)發(fā)的 3450 指南針系統(tǒng)。（ 2）基于模式識(shí)別的故障診斷 。為了滿足故障診斷的特殊需要，故障特征提取和分析技術(shù)正在經(jīng)歷，從時(shí)間領(lǐng)域分析到傅里葉頻域分析，從線性平穩(wěn)信號(hào)分析到非線性非平穩(wěn)分析，從頻域分析到時(shí)頻分析的過(guò)程。本特利內(nèi)華達(dá)公司也進(jìn)行了一系列實(shí)驗(yàn)研究轉(zhuǎn)子 軸承系統(tǒng)的故障機(jī)制 [9]。通過(guò)結(jié)合歷史數(shù)據(jù)，它可以定量的識(shí)別在目前條件下的關(guān)鍵部件，預(yù)測(cè)即將發(fā)生的異常和故障，并且預(yù)測(cè)它們未來(lái)的發(fā)展趨勢(shì)。本文討論了在機(jī)械狀態(tài)監(jiān)測(cè)與 故障診斷的最新進(jìn)展。 (2) the fault diagnosis based on pattern recognition。 2020. [7] Sohre JS. Troubleshooting to stop vibration of centrifugal. Petrop Chem. Engineer 1968。 169。 在這一領(lǐng)域一個(gè)簡(jiǎn)明的研究評(píng)論已經(jīng)被提出 [5， 6]。 Chen 等人 [12]利用非線性動(dòng)力學(xué)理論來(lái)分析了發(fā)電機(jī)軸振動(dòng)問(wèn)題的關(guān)鍵。然而，傳統(tǒng)的頻譜分析也有明顯的劣勢(shì)。因?yàn)樗鬁?zhǔn)確的系統(tǒng)模型，這種方法對(duì)于實(shí)踐中的復(fù)雜設(shè)備在經(jīng)濟(jì)上是不可行的。因此故障機(jī)制研究、信號(hào)處理和特征采集、故障成因研究和設(shè)備發(fā)展將更加緊密地聯(lián)系在一起，才能在將來(lái)實(shí)現(xiàn)故障診斷 專(zhuān)家系統(tǒng)。 IRD 公司開(kāi)發(fā)的 5911 系統(tǒng)，日本三菱公司開(kāi)發(fā)的 MHM 系統(tǒng)，丹麥 Bamp。根據(jù)他們所屬的主體系統(tǒng)，故障診斷，可分為三類(lèi)：（ 1）基于控制模型的故障診斷 ?？梢酝ㄟ^(guò)信號(hào)分析方法獲得一定的故障特征信息，然后可以做出相應(yīng)的故障診斷。之后，在上世紀(jì) 60 年代至 70 年代期間Shiraki [8]在日本對(duì)于故障機(jī)理的研究工作做了很大貢獻(xiàn)，并總結(jié)了豐富的現(xiàn)場(chǎng)故障排除經(jīng)驗(yàn)，以支持故障機(jī)制的理論。 機(jī)械設(shè)備故障診斷技術(shù)使用監(jiān)控機(jī)械運(yùn)轉(zhuǎn)和固定分析和提取重要特征的測(cè)量值來(lái)校準(zhǔn)關(guān)鍵部件的狀態(tài)。因此，即將到來(lái)的機(jī)械故障的識(shí)別系統(tǒng)，是防止系統(tǒng)故障的關(guān)鍵。 Signal processing 山東交通學(xué)院畢業(yè)設(shè)計(jì) 3 1. Introduction With the development of modern science and technology, machinery and equipment functions are being more and more perfect, and the machinery structure bees more largescale, integrated, intelligent and plicated. As a result, the ponent number increases significantly and the precision requirement for the part mating is stricter. The possibility and category of the related ponent failures therefore increase greatly. Malignant accidents caused by ponent faults occur frequently all over the world, and even a small mechanical fault may lead to serious consequences. Hence, efficient incipient fault detection and diagnosis are critical to machinery normal running. Although optimization techniques have been carried out in the machine design procedure and the manufacturing procedure to improve the quality of mechanical products, mechanical failures are still difficult to avoid due to the plexity of modern equipments. The condition monitoring and fault diagnosis based on advanced science and technology acts as an efficient mean to forecast potential faults and reduce the cost of machine malfunctions. This is the socalled mechan