油氣井后處理生產(chǎn)應(yīng)急智能監(jiān)控系統(tǒng)研究
本文選題:安全生產(chǎn) + 預(yù)測預(yù)警。 參考:《北京郵電大學(xué)》2013年碩士論文
【摘要】:本文首先闡述了凈化廠的基本情況,包括生產(chǎn)設(shè)備、工藝流程、DCS系統(tǒng),然后根據(jù)生產(chǎn)流程中各個采集數(shù)據(jù)點的關(guān)系,將數(shù)據(jù)點分為幾個不同的模塊,每個模塊可以單獨表示生產(chǎn)工藝中的一個流程。之后,針對項目建設(shè)的需求提出了對預(yù)警系統(tǒng)的設(shè)計思路,同時介紹了作為系統(tǒng)構(gòu)架平臺的ECM,提出了模型構(gòu)建和數(shù)據(jù)處理的基本流程和方法。 本文的主要研究對象是對預(yù)測預(yù)警平臺系統(tǒng)的構(gòu)建,因此本文隨后研究數(shù)據(jù)分析的各種方法的理論知識,確定了PCA為系統(tǒng)中需要應(yīng)用的主要的分析方法,并重點介紹了PCA算法的計算流程,另外,對于PLS、神經(jīng)網(wǎng)絡(luò)、小波、EWMA等分析方法也做了總體的概述。之后,提出了針對脫硫、脫水、硫磺回收、蒸氣四個生產(chǎn)單元的模型設(shè)計方法。 最后,在ECM中以主成分分析法PCA為主要模塊建模,建立了脫硫、脫水、硫磺回收、蒸汽等生產(chǎn)單元的故障預(yù)測預(yù)警的模型,并且對模型中的每個元件進行了詳細的介紹;在壓縮機的模型中,由于開關(guān)機狀態(tài)的分辨比較復(fù)雜,所以提出了以時間平移為核心的算法來確定機器的開關(guān)機狀態(tài)。通過建模后組成的系統(tǒng)研究長壽分廠設(shè)備運行的狀態(tài),保證了監(jiān)測對象的安全生產(chǎn)。 本項目實現(xiàn)了對凈化廠生產(chǎn)過程的實時監(jiān)控,使得在生產(chǎn)中的可能發(fā)生危險之前及時發(fā)現(xiàn)潛在的問題并排除成為了可能,為凈化廠安全生產(chǎn)提供了保障。
[Abstract]:This paper first describes the basic situation of the purification plant, including the production equipment, process flow and DCS system, then according to the relationship between the data points collected in the production process, the data points are divided into several different modules. Each module can represent a single process in the production process. After that, the design idea of early warning system is put forward according to the requirement of project construction. At the same time, the ECM, as the platform of system architecture, is introduced, and the basic flow and method of model construction and data processing are put forward. The main research object of this paper is the construction of the prediction and early warning platform system, so this paper then studies the theoretical knowledge of various methods of data analysis, and determines PCA as the main analysis method that needs to be applied in the system. The calculation flow of PCA algorithm is introduced in detail. In addition, the analysis methods such as PLS, neural network and wavelet EWMA are also summarized. After that, the model design method for desulfurization, dehydration, sulfur recovery and steam production unit is put forward. Finally, the principal component analysis (PCA) is used as the main module in ECM to model the failure prediction and warning model of desulfurization, dehydration, sulfur recovery, steam and other production units, and each component of the model is introduced in detail. In the compressor model, due to the complexity of the state resolution of the switch machine, an algorithm based on time translation is proposed to determine the state of the machine switch machine. After modeling, the system is used to study the operation state of the plant equipment, which ensures the safety of the monitoring object. The project realizes the real-time monitoring of the production process of the purification plant, which makes it possible to find the potential problems and eliminate the potential problems in time before the possible danger in the production, thus providing a guarantee for the safe production of the purification plant.
【學(xué)位授予單位】:北京郵電大學(xué)
【學(xué)位級別】:碩士
【學(xué)位授予年份】:2013
【分類號】:TP277;TE687
【參考文獻】
相關(guān)期刊論文 前10條
1 姚志強;周曦;戴蓓劏;;文本無關(guān)說話人識別中一種改進的模型PCA變換方法[J];電子與信息學(xué)報;2007年02期
2 吉明明;趙選民;唐偉廣;;可變抽樣區(qū)間的非正態(tài)EWMA均值控制圖[J];系統(tǒng)工程;2006年11期
3 王世f^;賀志國;;基于PCA特征的快速SAR圖像目標(biāo)識別方法[J];國防科技大學(xué)學(xué)報;2008年03期
4 劉韜;田洪祥;郭文勇;;主成分分析在某型柴油機光譜數(shù)據(jù)分析中的應(yīng)用[J];光譜學(xué)與光譜分析;2010年03期
5 印勇;何文娟;郭之強;郭攀;徐亦達;;分塊PCA和奇異值分解相結(jié)合的人臉識別算法[J];重慶大學(xué)學(xué)報;2012年08期
6 薛麗;;基于田口質(zhì)量損失函數(shù)的非正態(tài)EWMA控制圖優(yōu)化設(shè)計[J];工業(yè)工程;2011年05期
7 劉仲;邢彬朝;陳躍躍;;一種面向多核處理器的高效并行PCA-SIFT算法[J];國防科技大學(xué)學(xué)報;2012年04期
8 袁哲俊,徐羽中,馬玉林;面向AMT生產(chǎn)環(huán)境的EWMA質(zhì)量控制圖[J];哈爾濱工業(yè)大學(xué)學(xué)報;2000年01期
9 許仙珍;謝磊;王樹青;;基于PCA混合模型的多工況過程監(jiān)控[J];化工學(xué)報;2011年03期
10 王心醉;李巖;郭立紅;肖永鵬;董寧寧;楊麗梅;;基于雙向PCA和K近鄰的人臉識別算法[J];解放軍理工大學(xué)學(xué)報(自然科學(xué)版);2010年06期
,本文編號:2087378
本文鏈接:http://www.sikaile.net/kejilunwen/anquangongcheng/2087378.html