本文選題：再制造　切入點：質(zhì)量控制　出處：《沈陽工業(yè)大學(xué)》2017年碩士論文

【摘要】：廢舊機(jī)床再制造作為我國循環(huán)經(jīng)濟(jì)、綠色經(jīng)濟(jì)的重要組成部分之一,對于實現(xiàn)社會的可持續(xù)發(fā)展有著重大意義。而再制造產(chǎn)品質(zhì)量是制約再制造企業(yè)發(fā)展的重要因素,同時,再制造系統(tǒng)存在許多的不確定性:廢舊毛坯服役工況的不確定性、本身質(zhì)量缺陷的不確定性以及組織結(jié)構(gòu)的不確性,導(dǎo)致廢舊機(jī)床再制造過程的質(zhì)量控制較為困難,使得再制造產(chǎn)品的質(zhì)量難以保證�；诖�,以廢舊機(jī)床再制造過程為研究對象,根據(jù)再制造過程的不確定性特點,對再制造過程質(zhì)量異常在線監(jiān)控、異常識別以及質(zhì)量異常診斷與調(diào)整進(jìn)行研究,主要研究內(nèi)容如下:(1)針對毛坯質(zhì)量分布不確定的問題,提出基于動態(tài)、非正態(tài)的EWMA控制圖的方法。根據(jù)非參數(shù)方法中的Wilcoxon秩和檢驗的理論知識,應(yīng)用秩統(tǒng)計量,獲得與樣本數(shù)據(jù)分布無關(guān)的統(tǒng)計量;在此基礎(chǔ)上,通過不斷移動控制圖數(shù)據(jù)窗口來更新在線觀測點,并利用得分函數(shù)獲得動態(tài)的光滑參數(shù),構(gòu)建面向動態(tài)、非正態(tài)分布的再制造過程質(zhì)量EWMA控制圖,實現(xiàn)動態(tài)再制造過程質(zhì)量的自適應(yīng)監(jiān)控。(2)針對再制造過程質(zhì)量樣本少、質(zhì)量異常識別困難等問題,提出基于CS分解貝葉斯的主元分析方法。通過將CS分解貝葉斯空間估計的思想,特征子空間矩陣引入傳統(tǒng)的PCA的方法,解決傳統(tǒng)PCA需要大樣本的問題,從而能實現(xiàn)在小樣本條件下再制造過程質(zhì)量異常識別。(3)針對再制造過程質(zhì)量異常與引發(fā)因素之間的不確定性,采用基于粗糙集理論的再制造過程質(zhì)量異常診斷與調(diào)整系統(tǒng)。利用粗糙集理論知識求得再制造過程質(zhì)量影響因素相對再制造過程質(zhì)量特征的重要度,再通過不同質(zhì)量因素之間重要度大小,對質(zhì)量異常進(jìn)行溯源,找出引發(fā)質(zhì)量異常的原因并進(jìn)行調(diào)節(jié),實現(xiàn)再制造過程的異常診斷與調(diào)整。(4)以廢舊TPX6113鏜床的導(dǎo)軌再制造過程為例,通過基于動態(tài)、非正態(tài)的EWMA控制圖方法能靈敏地識別出再制造過程質(zhì)量異常,出現(xiàn)質(zhì)量異常后,基于改進(jìn)的PCA質(zhì)量異常識別模型能迅速識別出質(zhì)量異常類型,再通過基于粗糙集的質(zhì)量診斷模型對再制造質(zhì)量異常診斷,診斷出質(zhì)量異常原因并對過程異常進(jìn)行調(diào)節(jié),結(jié)果表明所提出的方法能夠解決再制造過程質(zhì)量控制問題,從而證明該方法的有效性和可行性。
[Abstract]:As one of the important parts of recycling economy and green economy in China, the remanufacture of waste machine tools is of great significance to the sustainable development of society.The quality of remanufacturing products is an important factor restricting the development of remanufacturing enterprises. At the same time, there are many uncertainties in the remanufacturing system: the uncertainty of service conditions of waste blank,The uncertainty of quality defects and the uncertainty of organization structure lead to the difficulty of quality control in the remanufacturing process of waste machine tools, which makes it difficult to guarantee the quality of remanufactured products.Based on this, taking the remanufacturing process of waste machine tool as the research object, according to the uncertainty characteristics of the remanufacturing process, this paper studies the on-line monitoring of the quality anomaly, the abnormal identification, the diagnosis and adjustment of the quality anomaly in the remanufacturing process.The main research contents are as follows: (1) to solve the problem of uncertain quality distribution of blank, a method based on dynamic and non-normal EWMA control chart is proposed.According to the theoretical knowledge of Wilcoxon rank sum test in nonparametric methods, the statistics independent of sample data distribution are obtained by using rank statistics, and the on-line observation points are updated by moving the control chart data window continuously.The dynamic smooth parameters are obtained by using the score function, and the EWMA control chart of the remanufacturing process quality oriented to dynamic and non-normal distribution is constructed, which realizes the self-adaptive monitoring of the quality of the dynamic remanufacturing process.This paper presents a principal component analysis method based on CS decomposition Bayes.Based on the idea of CS decomposition Bayesian space estimation, the characteristic subspace matrix is introduced into the traditional PCA method to solve the problem that traditional PCA needs large samples.So it can realize the recognition of remanufacturing process quality anomaly under the condition of small sample. Aiming at the uncertainty between the remanufacturing process quality anomaly and the trigger factor, a rough set theory based remanufacturing process quality anomaly diagnosis and adjustment system is adopted.By using rough set theory, the importance of the quality influencing factors in the remanufacturing process relative to the quality characteristics of the remanufacturing process is obtained, and then the source of the quality anomalies is traced through the importance degree between the different quality factors.Find out the cause of the abnormal quality and adjust it to realize the abnormal diagnosis and adjustment of the remanufacturing process. Take the remanufacturing process of the used TPX6113 boring machine as an example, based on the dynamic,The non-normal EWMA control chart method can sensitively identify the quality anomaly in the remanufacturing process. After the quality anomaly occurs, the improved PCA quality anomaly recognition model can quickly identify the type of the quality anomaly.Then, by using the quality diagnosis model based on rough set to diagnose the remanufacturing quality anomaly, the cause of the quality anomaly is diagnosed and the process anomaly is adjusted. The results show that the proposed method can solve the problem of the remanufacturing process quality control.Thus, the effectiveness and feasibility of the method are proved.
【學(xué)位授予單位】：沈陽工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：F426.4;F713.2;F273.2

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 黃友朋;趙山;許凡;方彥軍;;EEMD排列熵與PCA-GK的滾動軸承聚類故障診斷[J];河南科技大學(xué)學(xué)報(自然科學(xué)版);2017年02期

2 許凡;方彥軍;張榮;;基于EEMD模糊熵的PCA-GG滾動軸承聚類故障診斷[J];計算機(jī)集成制造系統(tǒng);2016年11期

3 李金艷;余忠華;徐宣國;;信息不完備情況下多因素工序質(zhì)量診斷方法[J];哈爾濱工業(yè)大學(xué)學(xué)報;2016年07期

4 徐濱士;朱勝;史佩京;;綠色再制造技術(shù)的創(chuàng)新發(fā)展[J];焊接技術(shù);2016年05期

5 石書彪;陳煥新;李冠男;胡云鵬;黎浩榮;胡文舉;;基于小波去噪和神經(jīng)網(wǎng)絡(luò)的冷水機(jī)組故障診斷[J];制冷學(xué)報;2016年01期

6 侯彥東;閆治宇;金勇;;小樣本下基于特征子空間估計的故障診斷算法[J];上海交通大學(xué)學(xué)報;2015年06期

7 王培良;夏春江;;基于PCA-PDBNs的故障檢測與自學(xué)習(xí)辨識[J];儀器儀表學(xué)報;2015年05期

8 李恩重;史佩京;徐濱士;鄭漢東;宋金鵬;;我國再制造政策法規(guī)分析與思考[J];機(jī)械工程學(xué)報;2015年19期

9 徐濱士;梁秀兵;史佩京;蔡志海;;我國再制造工程及其產(chǎn)業(yè)發(fā)展[J];表面工程與再制造;2015年02期

10 周帆;江志剛;張華;張楚鋒;張旭剛;;基于知識的廢舊機(jī)床再制造設(shè)計方法研究[J];制造技術(shù)與機(jī)床;2014年12期

，

本文編號：1724072