流程工業(yè)過程故障檢測的特征提取方法研究
發(fā)布時間：2018-08-08 17:44
【摘要】：機(jī)器的出現(xiàn)使生產(chǎn)效率和生產(chǎn)安全得到了提升,但機(jī)械設(shè)備的故障同樣也會導(dǎo)致財(cái)產(chǎn)損失和人員傷亡。隨著工業(yè)過程不斷向著大型化、復(fù)雜化的趨勢發(fā)展,傳統(tǒng)的依靠對每個變量實(shí)施獨(dú)立監(jiān)控的方法不再可行,并且建立精確的解析模型也越來越困難。與此同時,隨著集散控制系統(tǒng)的廣泛應(yīng)用,從多個角度反映系統(tǒng)運(yùn)行狀態(tài)的大量數(shù)據(jù)被記錄了下來。于是,為了進(jìn)一步提高過程監(jiān)控的效率,數(shù)據(jù)驅(qū)動的故障檢測方法開始興起。 實(shí)際工業(yè)過程通常有多種復(fù)雜的特性,如多工況、時變、復(fù)雜數(shù)據(jù)分布等,而在最初進(jìn)行研究時,人們會通過設(shè)定一些假設(shè)條件如過程運(yùn)行在單一穩(wěn)定工況下,變量服從高斯分布等,以此來簡化研究問題。但是隨著研究工作的深入開展,就需要克服這些假設(shè),研究適用性更強(qiáng)的故障檢測方法。本文旨在針對工業(yè)過程中的復(fù)雜數(shù)據(jù)分布、多工況和時變問題,通過利用數(shù)據(jù)局部和全局結(jié)構(gòu)信息,提出新的特征提取方法,進(jìn)一步提高故障檢測模型的精度。具體內(nèi)容如下： (1)針對單工況下訓(xùn)練數(shù)據(jù)具有復(fù)雜數(shù)學(xué)分布的問題,提出了一種基于局部線性嵌入(Locally linear embedding, LLE)和支持向量數(shù)據(jù)描述(Support vector data description, SVDD)的故障檢測算法。該算法充分利用了LLE在特征提取和SVDD在建立統(tǒng)計(jì)量時均不受數(shù)據(jù)數(shù)學(xué)分布約束的優(yōu)勢,所以可以建立精確的監(jiān)控模型。此外,利用最小二乘回歸策略,解決了LLE不能得到原始空間到特征空間投影矩陣的問題,保證了在線監(jiān)控的效率。 (2)針對傳統(tǒng)監(jiān)控方法在特征提取時會破壞原始數(shù)據(jù)局部或全局結(jié)構(gòu)的問題,提出了一種局部-非局部嵌入(Local and nonlocal embedding, LNLE)算法。LNLE利用LLE的目標(biāo)函數(shù)約束樣本與其近鄰點(diǎn)間的相對位置,同時設(shè)計(jì)了一個新的目標(biāo)函數(shù)約束樣本與其非近鄰點(diǎn)間的相對位置,最終通過求解雙重優(yōu)化問題,將原始數(shù)據(jù)的局部和全局結(jié)構(gòu)信息完整的保存在特征空間中,進(jìn)一步降低了特征提取過程中的信息損失。 (3)針對訓(xùn)練數(shù)據(jù)來自多個生產(chǎn)工況即訓(xùn)練數(shù)據(jù)的數(shù)學(xué)分布具有多峰性的問題,提出了一種協(xié)調(diào)混合因子分析(Aligned mixture factor analysis, AMFA)算法。AMFA在傳統(tǒng)多模型法的基礎(chǔ)上,通過把樣本在特征空間中的表達(dá)具有唯一性作為優(yōu)化過程的約束條件,將各個局部模型協(xié)調(diào)整合得到一個全局模型,從而使監(jiān)控模型不僅包含每個工況的獨(dú)有特征而且包含工況間數(shù)據(jù)的相關(guān)性信息。同時,在線監(jiān)控時,由于不需要判斷新樣本屬于何種工況或應(yīng)該使用哪個局部模型進(jìn)行故障檢測,所以監(jiān)控效率得到了提升。 (4)針對現(xiàn)有聚類算法在處理多工況訓(xùn)練數(shù)據(jù)時無法自動求得工況數(shù)以及部分聚類算法可能陷入局部最優(yōu)的問題,提出了一種工業(yè)過程數(shù)據(jù)聚類算法。該算法利用了樣本的時序相關(guān)性和工業(yè)數(shù)據(jù)中同類數(shù)據(jù)總是相鄰的特點(diǎn),通過尋找擴(kuò)展矩陣中的“斷裂點(diǎn)”,提高了聚類的效率和精度。此外,在將局部模型整合為全局模型時,通過同時約束樣本與其近鄰點(diǎn)及非近鄰點(diǎn)間的相對位置,并引入權(quán)重系數(shù)平衡二者的比例,在特征提取過程中更完整的保存了原始數(shù)據(jù)的局部和全局結(jié)構(gòu)。 (5)針對同時具有時變、多工況和復(fù)雜數(shù)據(jù)分布的工業(yè)過程,提出了移動窗局部離群因子和移動窗局部離群概率兩種故障檢測方法。利用兩種方法基于近鄰點(diǎn)計(jì)算局部密度從而不受復(fù)雜數(shù)據(jù)分布影響的優(yōu)勢,保證了監(jiān)控模型的精度。同時,分別針對兩種算法提出了相應(yīng)的移動窗快速更新策略,提高了模型更新速度,保證了監(jiān)控效率。而對于工況變化過程,又提出了半監(jiān)督的模型切換機(jī)制,通過短時間內(nèi)強(qiáng)制接受每個新樣本使監(jiān)控模型能迅速跟蹤工況的變化。為了減小將故障或擾動更新到窗口中的概率,分別針對兩種算法提出了“盲更新”的終止條件,保證了算法的持續(xù)有效性。 本文在對上述方法進(jìn)行理論分析的同時,在數(shù)值例子、非等溫連續(xù)式攪拌釜(Non-isothermal continuous stirred tank reactor, CSTR)和田納西伊斯曼(Tennessee eastman,TE)過程的仿真框架下,通過設(shè)計(jì)不同的測試情景并與文獻(xiàn)中類似的方法進(jìn)行對比,驗(yàn)證了本文提出算法的有效性和實(shí)用性。
[Abstract]:The emergence of machines makes production efficiency and production safety improved, but mechanical equipment failures also lead to property losses and casualties. As the industrial process continues to become larger and more complex, the traditional method of independent monitoring of each variable is no longer feasible and an accurate analytical model is established. At the same time, with the wide application of distributed control system, a large number of data which reflect the running state of the system from many angles have been recorded. So, in order to further improve the efficiency of process monitoring, the data driven fault detection method began to rise.
The actual industrial process usually has a variety of complex characteristics, such as multiple working conditions, time-varying, complex data distribution, and so on. In the initial research, people will simplify the research problem by setting some assumptions, such as the process running in a single stable condition, the variable obeys the Gauss distribution, so as to simplify the research problem. We need to overcome these hypotheses and study more applicable fault detection methods. This paper aims at the complex data distribution in the industrial process, multi condition and time-varying problem. By using the local and global information of the data, a new feature extraction method is proposed to further improve the accuracy of the obstacle detection model.
(1) a fault detection algorithm based on Locally linear embedding (LLE) and support vector data description (Support vector data description, SVDD) is proposed to solve the problem of complex mathematical distribution of training data under single operating conditions. This algorithm takes advantage of LLE in feature extraction and SVDD in the establishment of statistics. In addition, the least squares regression strategy is used to solve the problem that LLE can't get the original space to the feature space projection matrix, and the efficiency of on-line monitoring is guaranteed.
(2) aiming at the problem that the traditional monitoring method can destroy the local or global structure of the original data when feature extraction, a local non local embedding (Local and nonlocal embedding, LNLE) algorithm.LNLE is used to use the phase pair position between the target function constraint sample and its near neighbor of the target function of LLE, and a new objective function constraint sample is designed. By solving the dual optimization problem, the local and global structure information of the original data is preserved in the feature space, and the information loss in the process of feature extraction is further reduced.
(3) aiming at the problem that the mathematical distribution of training data from multiple production conditions is multi peak, a coordinated mixed factor analysis (Aligned mixture factor analysis, AMFA) algorithm.AMFA is proposed on the basis of the traditional multi model method, by which the expression of the sample in the feature space is unique as the optimization process. The constraints of each local model are coordinated and integrated to get a global model, so that the monitoring model not only contains the unique features of each working condition but also contains the correlation information between the working conditions and the data. At the same time, in the online monitoring, it does not need to judge what working conditions of the new sample or which local model should be used for fault detection. So the monitoring efficiency has been improved.
(4) in view of the problem that the existing clustering algorithm can not automatically get the number of working conditions and the partial clustering algorithm may fall into local optimum when processing the multi condition training data, an industrial process data clustering algorithm is proposed. The algorithm uses the temporal correlation of the sample and the characteristics of the similar data of the same industry in the industrial data. The "breaking point" in the extended matrix improves the efficiency and accuracy of clustering. In addition, when the local model is integrated into a global model, the proportion of the two persons is balanced by the relative position between the constraint samples and their nearest neighbors and the non nearest neighbors, and the original data are preserved more completely during the process of feature extraction. The Department and the global structure.
(5) aiming at the industrial process with time-varying, multi working conditions and complex data distribution, two fault detection methods are proposed for local outliers of mobile windows and local outlier probability of moving windows. The accuracy of the monitoring model is ensured by using two methods to calculate the local density based on the nearest neighbor and thus not affected by the complex data distribution. In order to improve the speed of updating the model and ensure the efficiency of monitoring, a semi supervised model switching mechanism is put forward for two kinds of algorithms, which can improve the speed of model updating and ensure the efficiency of monitoring. The probability of failure or disturbance updating to the window is analyzed, and the termination condition of blind updating is proposed for the two algorithms, which guarantees the continual validity of the algorithm.
In this paper, in the theoretical analysis of the above methods, in the simulation framework of numerical examples, non isothermal continuous stirred kettle (Non-isothermal continuous stirred tank reactor, CSTR) and Tennessee Eastman (Tennessee Eastman, TE) process, different test scenarios are designed and compared with similar methods in the literature. The validity and practicability of the algorithm are proposed in this paper.
【學(xué)位授予單位】：華東理工大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2015
【分類號】：TH165.3
本文編號：2172540
[Abstract]:The emergence of machines makes production efficiency and production safety improved, but mechanical equipment failures also lead to property losses and casualties. As the industrial process continues to become larger and more complex, the traditional method of independent monitoring of each variable is no longer feasible and an accurate analytical model is established. At the same time, with the wide application of distributed control system, a large number of data which reflect the running state of the system from many angles have been recorded. So, in order to further improve the efficiency of process monitoring, the data driven fault detection method began to rise.
The actual industrial process usually has a variety of complex characteristics, such as multiple working conditions, time-varying, complex data distribution, and so on. In the initial research, people will simplify the research problem by setting some assumptions, such as the process running in a single stable condition, the variable obeys the Gauss distribution, so as to simplify the research problem. We need to overcome these hypotheses and study more applicable fault detection methods. This paper aims at the complex data distribution in the industrial process, multi condition and time-varying problem. By using the local and global information of the data, a new feature extraction method is proposed to further improve the accuracy of the obstacle detection model.
(1) a fault detection algorithm based on Locally linear embedding (LLE) and support vector data description (Support vector data description, SVDD) is proposed to solve the problem of complex mathematical distribution of training data under single operating conditions. This algorithm takes advantage of LLE in feature extraction and SVDD in the establishment of statistics. In addition, the least squares regression strategy is used to solve the problem that LLE can't get the original space to the feature space projection matrix, and the efficiency of on-line monitoring is guaranteed.
(2) aiming at the problem that the traditional monitoring method can destroy the local or global structure of the original data when feature extraction, a local non local embedding (Local and nonlocal embedding, LNLE) algorithm.LNLE is used to use the phase pair position between the target function constraint sample and its near neighbor of the target function of LLE, and a new objective function constraint sample is designed. By solving the dual optimization problem, the local and global structure information of the original data is preserved in the feature space, and the information loss in the process of feature extraction is further reduced.
(3) aiming at the problem that the mathematical distribution of training data from multiple production conditions is multi peak, a coordinated mixed factor analysis (Aligned mixture factor analysis, AMFA) algorithm.AMFA is proposed on the basis of the traditional multi model method, by which the expression of the sample in the feature space is unique as the optimization process. The constraints of each local model are coordinated and integrated to get a global model, so that the monitoring model not only contains the unique features of each working condition but also contains the correlation information between the working conditions and the data. At the same time, in the online monitoring, it does not need to judge what working conditions of the new sample or which local model should be used for fault detection. So the monitoring efficiency has been improved.
(4) in view of the problem that the existing clustering algorithm can not automatically get the number of working conditions and the partial clustering algorithm may fall into local optimum when processing the multi condition training data, an industrial process data clustering algorithm is proposed. The algorithm uses the temporal correlation of the sample and the characteristics of the similar data of the same industry in the industrial data. The "breaking point" in the extended matrix improves the efficiency and accuracy of clustering. In addition, when the local model is integrated into a global model, the proportion of the two persons is balanced by the relative position between the constraint samples and their nearest neighbors and the non nearest neighbors, and the original data are preserved more completely during the process of feature extraction. The Department and the global structure.
(5) aiming at the industrial process with time-varying, multi working conditions and complex data distribution, two fault detection methods are proposed for local outliers of mobile windows and local outlier probability of moving windows. The accuracy of the monitoring model is ensured by using two methods to calculate the local density based on the nearest neighbor and thus not affected by the complex data distribution. In order to improve the speed of updating the model and ensure the efficiency of monitoring, a semi supervised model switching mechanism is put forward for two kinds of algorithms, which can improve the speed of model updating and ensure the efficiency of monitoring. The probability of failure or disturbance updating to the window is analyzed, and the termination condition of blind updating is proposed for the two algorithms, which guarantees the continual validity of the algorithm.
In this paper, in the theoretical analysis of the above methods, in the simulation framework of numerical examples, non isothermal continuous stirred kettle (Non-isothermal continuous stirred tank reactor, CSTR) and Tennessee Eastman (Tennessee Eastman, TE) process, different test scenarios are designed and compared with similar methods in the literature. The validity and practicability of the algorithm are proposed in this paper.
【學(xué)位授予單位】：華東理工大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2015
【分類號】：TH165.3
【參考文獻(xiàn)】
中國期刊全文數(shù)據(jù)庫 前10條
1 胡益;馬賀賀;侍洪波;;On-Line Batch Process Monitoring Using Multiway Kernel Partial Least Squares[J];Journal of Donghua University(English Edition);2011年06期
2 王洪江;孫保民;田進(jìn)步;;定性仿真在鍋爐狀態(tài)監(jiān)控和故障診斷中的應(yīng)用[J];工程熱物理學(xué)報(bào);2007年01期
3 陳國金,梁軍,錢積新;獨(dú)立元分析方法(ICA)及其在化工過程監(jiān)控和故障診斷中的應(yīng)用[J];化工學(xué)報(bào);2003年10期
4 葛志強(qiáng);宋執(zhí)環(huán);;一種新的多工況過程在線監(jiān)測方法[J];化工學(xué)報(bào);2008年01期
5 張少捷;王振雷;錢鋒;;基于LTSA的FS-SVDD方法及其在化工過程監(jiān)控中的應(yīng)用[J];化工學(xué)報(bào);2010年08期
6 王凌,王雄;流程工業(yè)CIMS體系結(jié)構(gòu)和生產(chǎn)執(zhí)行系統(tǒng)[J];計(jì)算機(jī)工程與應(yīng)用;2003年10期
7 柴天佑,金以慧,任德祥,邵惠鶴,錢積新,李平,桂衛(wèi)華,鄭秉霖;基于三層結(jié)構(gòu)的流程工業(yè)現(xiàn)代集成制造系統(tǒng)[J];控制工程;2002年03期
8 許建新;侯忠生;;數(shù)據(jù)驅(qū)動系統(tǒng)方法概述(英文)[J];自動化學(xué)報(bào);2009年06期
9 周東華;胡艷艷;;動態(tài)系統(tǒng)的故障診斷技術(shù)[J];自動化學(xué)報(bào);2009年06期
10 常玉清;王姝;譚帥;王福利;楊潔;;基于多時段MPCA模型的間歇過程監(jiān)測方法研究[J];自動化學(xué)報(bào);2010年09期
，本文編號：2172540
本文鏈接：http://sikaile.net/jixiegongchenglunwen/2172540.html
最近更新
教材專著
